| Bulletin of
_ The Natural History Museum
| Pee 18 JUN 1998
| : PRESENTED
} GENERAL LIBRARY
|
|
f
:
(
| A)
\ aed
| THE
' NATURAL
| HISTORY
i MUSEUM
!
| |
fi
yf hi}
The Bulletin of The Natural History Museum (formerly: Bulletin of the British Museum (Natural History) ), instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology.
The Entomology Series is produced under the editorship of the Keeper of Entomology: R.I. Vane-Wright Editor of Bulletin: Dr Gaden S. Robinson
Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff and by specialists from elsewhere who make use of the Museum’s resources. Many of the papers are works of reference that will remain indispensable for years to come. All papers submitted for publication are subjected to external peer review before acceptance.
aye volume contains about 192 pages, made up by two numbers, published in the Spring and Autumin. Subscriptions may be placed for one or more of the series on an annual basis. Individual numbers and back numbers can be purchased and a Bulletin catalogue, by series, is available. Orders and enquiries should be sent to:
Intercept Ltd.
P.O. Box 716
Andover
Hampshire SP10 1YG Telephone: (01264) 334748 Fax: (01264) 334058
Claims for non-receipt of issues of the Bulletin will be met free of charge if received by the Publisher within 6 months for the UK, and 9 months for the rest of the world.
World List abbreviation: Bull. nat. Hist. Mus. Lond. (Ent.)
© The Natural History Museum, 1998
Entomology Series ISSN 0968-0454 Vol. 67, No. 1, pp. 1-152
The Natural History Museum Cromwell Road London SW7 5BD Issued 25 June 1998
Typeset by Ann Buchan (Typesetters), Middlesex Printed in Great Britain by Henry Ling Ltd, at the Dorset Press, Dorchester, Dorset
Bull. nat. Hist. Mus. Lond. (Ent.) 67(1): 1-64 Issued 25 June 1998
Mealybugs of the genera Eumyrmococcus Silvestri and Xenococcus Silvestri associated with the ant genus Acropyga Roger and a review of the subfamily Rhizoecinae (Hemiptera, Coccoidea, Pseudococcidae)
THE NATURAL HISTORY MUSEUM
18 JUN 1998
D.J. WILLIAMS__ GENERAL LIBRARY
Department of Entomology, The Natural History Museum, Cromwell Road, London
SW7 SBD. CONTENTS PATIO GICH OM) sxe ceeussscaseivdsceasvene caanseasissesciaiassnsscesosdeveteegecsdetepar doneorae cack coSeResene aaayeGs cane teensaesecesees 2 Mite nral AM GITMC lM OGS: 1. .tac. vs cncancus-taengascanssancis+sas <deeemee ae oeee eee renter Woe naeacee ety hamper ge tas a scan se 2 FALE VMAL ONS OUNMSMUUM OMS. « excesses ccccrsanacsnszegue «-cgnsaceteesseacedtascera shod: teseoeeesesaneminet meee nace leeacnoeee 3) AN GRITOWHECPEIICIES ccctascvcccssovnsussssanesusscvesadcunsesasasecnseaunmancaene ax seubesvon une oetisertedss sees adanmsvaaetaverdeetwase 3 Feed Oo trea CME COMMON CHINN OU LANCE Yes y cca ge per esc cet -«~tnwnc Soba tvavae scenes ec sete eal raunvans Sate tates ata tana ap ae 3 Life cycles Parasitism Morphology of Eumyrmococcus and Xenococcus (adult females) GreMelAAI SPECIES recreate acca Macste tect swereness do esaua anasais aig anes be amuse naeniy eras duusieiabiicessemenisn Z EEE BITING O GGL SRC ee nett Fatt OTe Nc SNe ea oer cc pc nae Rando tes ces do ayer cass oh canauct Ve eNaas vwecds nas « 8 Kee VaLOISHE CIES OM My ITIOGOCOMS (ACUIT LEINALCS) 0 ccearcaccones anche ceearcacccntcncscrtaccvonsyotscnrnacsns 9 Key to species of Xenococcus (adult females) Review omreneraimitte MZ OC CINAC (i eerectecs..coeenceqreanearseeocseseseeresl Key to genera of the subfamily Rhizoecinae (adult females) ............csceeceseeseseeeeseeseneeseeees 30 List of Rhizoecinae associated with Acropyga with discussion of specificity of the IME ALY DUP ANE ASSOCLALIONS. \sescesecsaesusvecsescnvonacsxtereeetreetosetts con tsecettacecn ace aav¥entecceresds si sttsteere 30 MRS Ke carntSc aoe cet tacn case anahonghaneascannannenhannssamare=s encod eeeteerertysecaranctee tee tenae Rec eenee acer ae vod octcate stot 64
SyNopsIS. The hypogeic mealybug genera Eumyrmococcus Silvestri and Xenococcus Silvestri, their distribution throughout southern Asia and Australasia, and their association with the ant genus Acropyga Roger are discussed. Eumyrmococcus is also discussed from Europe and South Africa. Descriptions or redescriptions of 17 species of Eumyrmococcus are presented and illustrated, including E. corinthiacus Williams, E. falciculosus sp. n., E. kolombangarae sp. n., E. kruiensis sp. n., E. kusiacus sp. n., E. lamondicus sp. n., E. lanuginosus sp. n., E. maninjauensis sp. n., E. neoguineensis sp. n., E. nipponensis Terayama, E. queenslandicus sp. n., E. recalvus sp. n., E. sarawakensis sp. n., E. scorpioides (De Lotto), E. smithii Silvestri, E. sulawesicus sp. n. and E. taylori sp. n.The genus Xenococcus includes two species, X. acropygae sp. n. andX. annandalei Silvestri. It is now known that X. annandalei is a local species from a small area in India. The female pupal instar, already known in Xenococcus, 1s also discussed inEumyrmococcus. Full page illustrations for each species and keys to species of Eumyrmococcus and Xenococcus are provided.
© The Natural History Museum, 1998
D.J. WILLIAMS
The position of Eumyrmococcus and Xenococcus within the subfamily Rhizoecinae is discussed and a review of all the genera included in the subfamily is provided. A new genus and species, Leptorhizoecus deharvengi, is described from Sumatra and the generic name Radicoccus
Hambleton is synonymised with Rhizoecus.
INTRODUCTION
In a list of family-group names of the Coccoidea, Williams (1969a) erected the tribe Rhizoecini, within the mealybug family Pseudococcidae, for Rhizoecus Kiinckel d’Herculais and a few related genera, all inhabiting soil, leaf litter, rotting logs or feeding on roots. Some of these genera are now known to be identical with others and some other genera have been added since. After a comprehensive study of the la- bium of the Coccoidea, Koteja (1974a, 1974b) accepted the group as the subfamily Rhizoecinae, mainly on the basis that the labium is very narrow. This group also lacks cerarii and the antennae are usually strongly geniculate, with never more than 6 antennal segments, the terminal segment always tapering and pointed. The subfamily status has been accepted by most workers. The other subfamilies Trabutininae, Pseudococcinae and Sphaerococcinae, discussed by Koteja, have also been accepted by many students of the Coccoidea although the status of each subfamily is still ill-de- fined.
The purpose of this work is to revise the genera Eumyrmococcus and Xenococcus. A few species of Eumyrmococcus had been studied already for the au- thor’s current work on the mealybugs of southern Asia. Some interesting collections, however, sent by Dr R.W. Taylor, CSIRO, Canberra, from Australasia, as- sociated with the ant genus Acropyga, have revealed some remarkable species. Any attempt to publish on southern Asian species without a knowledge of all the others would be undesirable. It is for this reason that a study is presented here of all the known species of Eumyrmococcus, now totalling 17.
It is clear that most of the records of the related genus Xenococcus have been based on a misconcep- tion. The type species X. annandalei Silvestri, appears to be a local species, at present living only on Barkuda I., Orissa Province, India. Another species, previously identified as X. annandalei from India, most of south- ern Asia and Australasia, is different, and is here described as new.
Eumyrmococcus and Xenococcus were also included in the subfamily Rhizoecinae by Tang (1992) who erected the tribe Xenococcini for them. According to Tang, this tribe differs from the Rhizoecini, the only other tribe, in lacking ostioles. The distinction may hold for Chinese and neighbouring species but not in the subfamily as a whole as known worldwide. It is difficult to separate Eumyrmococcus and Xenococcus, for instance, without reference to Neochavesia
Williams & Granara de Willink, a South American genus. This study of Eumyrmococcus and Xenococcus is followed, therefore, by a discussion of all genera of the subfamily Rhizoecinae. Included genera accepted here are Rhizoecus, Capitisetella Hambleton, Pseudorhizoecus Green, Geococcus Green, Leptorhizoecus gen. nov., Pygmaeococcus McKenzie, Neochavesia, Eumyrmococcus and Xenococcus. A key to these genera is also included.
Few specimens of Eumyrmococcus have been avail- able for this study. Collecting these mealybugs is not easy and acknowledgement is always due to the col- lectors who have searched for them in ants’ nests or have taken the trouble to extract them from soil sam- ples. Species records represent isolated collections over a wide area. All the species are minute and their preparation on to microscope slides is often tedious. Distinguishing adults from immatures preserved in vials is almost impossible and it is often disappointing to find only one or two adult females in any sample. Even more disappointing is the complete absence of adults.
The present study of Eumyrmococcus and Xenococcus must be recognized as a preliminary iden- tification guide and not a phylogenetic study. Some possible species-groups can be distinguished and these are discussed in the introduction to the section on “Species and Genera’ (p. 7). Itis hoped that the present study of these mealybugs and their ant relationships will form the basis for a future cladistic study. Such a study could also be extended to the other rhizoecine genera discussed here, at least to some of the monotypic genera which need further investigation.
MATERIAL AND METHODS
The slide-mounting techniques and the methods of illustration described by Williams & Watson (1988) and Williams & Granara de Willink (1992) are fol- lowed. The terminology in these works is adopted here except for the terms bitubular and tritubular pores. In the present work they are referred to as bitubular cerores and tritubular cerores, terms that were widely used by Hambleton (1976) and adopted from MacGillivray (1921).
Measurements of the body length and width are in millimetres but all other measurements, even if larger than the body length, are given in microns.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 3
ABBREVIATIONS OF INSTITUTIONS
The following acronyms are used throughout for the depositories of specimens.
ANIC Australian National Insect Collection, Canberra, Australia.
BMNH _ The Natural History Museum, London, UK.
IEAUN Istituto di Entomologia Agraria dell’ Universita di Napoli, Portici, Italy.
MNHN- Muséum National d’ Histoire Naturelle, Paris, France.
NIAES _ National Institute of Agro-Environmental
Science, Tsukuba-shi, Japan. QM Queensland Museum, South Brisbane, Queens- land, Australia.
SANC National Collection of Insects, Pretoria, South Africa.
USNM _ National Museum of Natural History [Scale insect Collection], Beltsville, Maryland, USA.
ACKNOWLEDGEMENTS
Much of the material discussed in this work was sent by Dr R.W. Taylor, CSIRO, Canberra, Australia, as part of his ant studies of southern Asia and Australasia. | am much indebted to Dr Taylor for the opportunity of examining this material. Daniele Matile-Ferrero, Muséum National d’Histoire Naturelle, Paris, France, kindly sent all the mealybug material extracted from soil samples collected by Drs Louis Deharveng and Anne Bedos, Université Paul Sabatier, Toulouse. Dr D.R. Miller, Systematic Entomology Laboratory, USDA, Beltsville, Maryland, USA, arranged the loan of material from the US National Museum of Natural History housed at Beltsville, and has kindly spent time examining specimens and giving useful information. Dr Mamoru Terayama, Department of Biology, College of Arts and Sciences, University of Tokyo, Japan, has generously sent material of Eumyrmococcus nipponensis Terayama and E. smithii Silvestri used in the descriptions for this work. lan M. Millar, Plant Protection Research Institute, Pretoria, South Africa, has kindly sent for study the holotype and paratypes of E. scorpioides (De Lotto). Dr Penny J. Gullan, Australian National University, Can- berra, Australia, read the entire draft manuscript, suggested improvements and offered perceptive criticism, which I have heeded throughout. I am much indebted to Dr Gullan for spending her time and showing keen interest in this work.
HABIT AND ECONOMIC IMPORTANCE
It has been been postulated by Koteja (1984, 1985) and well summarised by Gullan & Kosztarab (1997) that the ancestors of present-day Coccoidea lived on the leaf-litter layer, feeding on dead and decaying matter,
or even on associated fungi and bacteria. Some soil- inhabiting coccoids are, apparently, secondarily inhabitants but Koteja suggests that Margarodes Guilding (Margarodidae) and Rhizoecus (Pseudococcidae) might have spent all of their phylogeny in soil or litter. The underground habit of Rhizoecus could not have been acquired secondarily from an aerial habit on stems, twigs and leaves (Koteja, 1985).
Endosymbionts of scale insects are diverse and may have been acquired via the alimentary canal during the original leaf-litter habit. Mealybugs may have even changed their feeding behaviour by living a symbiotic relationship with ants (Tremblay, 1989) but the endosymbionts of most mealybugs, although differ- ent, are nevertheless related (Buchner, 1969). The symbionts of Puto Signoret and Macrocerococcus Leonardi are different from others in the Pseudococcidae and those of Rastrococcus Ferris are so distinct that Rastrococcus should be separated from the Pseudococcidae (Buchner, 1957, 1969). Tremblay (1977) recognizes two main symbiotic adaptations among the pseudococcids, the Pseudococcus Westwood type and the Puto type. No symbionts of Rhizoecus and its close relatives have been studied but Silvestri (1924) showed that Xenococcus has a pseudococcid type of endosymbiosis. Buchner (1957, 1969) apparently studied symbionts of Eumyrmococcus (without mentioning the species) and commented that these are a pseudococcid type, near those of Phenacoccus Cockerell.
Rhizoecinae live underground, mostly feeding on rootlets. Many species have only been collected from Berlese funnel apparatus but many are known from leaf litter and rotting logs. The only aerial habit re- ported is that of Rhizoecus mexicanus (Hambleton) on leaves of Zygocactus truncatus; otherwise this mealybug normally feeds on cactus roots (Hambleton, 1979):
Species of many genera of Rhizoecinae are associ- ated with ants of the genus Acropyga Roger which attend the mealybugs and feed on their excreted hon- eydew. The association may be an old one because some of these ant species are known to be obligate coccidophiles and many of the mealybug species are probably obligate myrmecophiles, although some mealybug species reported with ants are capable of living without them. Rhizoecus coffeae Laing is a pest of coffee in Brazil where it lives on the roots in association with Acropyga (Rhizomyrma) paramibensis Borgmeier (Biinzli, 1935). The mealybug is well protected from excessive change in dryness and wetness caused by the weather because the nests are well drained and the mealybugs are capable of moving to suitable feeding sites. Accord- ing to Biinzli, the situation is different from that of the ant-free mealybugs. Flanders (1957) reported that
4
the myrmecophilous mealybug Eumyrmococcus (= Neochavesia) sp. only feeds on roots exposed by the ant A. (Rhizomyrma) fuhrmanni Forel and that the number of mealybugs feeding at any one time is regulated by the ant.
Ants associated with Eumyrmococcus and Xenococcus are also species of Acropyga and these mealybugs have been collected within the ants’ nests, in ant tunnels, or from the mandibles of flying queens. A few Eumyrmococcus species described here as new, have been collected without ant data or simply labelled ‘with ants’ but it is reasonable to suggest that all these mealybugs may be associated with Acropyga (see list of Acropyga-Rhizoecinae associations p. 30).
When carrying mealybugs in the mandibles, ants grip species of Rhizoecus and Geococcus at a point near the metathorax (Biinzli, 1935). Ants apparently grasp individuals of Eumyrmococcus at the dilated cephalothorax (Roepke 1930; Prins 1982; Buschinger et al., 1987).
Ants of the genus Acropyga are discussed here in the four subgenera listed by Bolton (1995a). Some of the unidentified species of Australasian Acropyga may be undescribed if not discussed by Taylor and Brown (1985) orTaylor (1992). Of the 56 species ofAcropyga, about half occur in the Neotropical Region and the remainder are distributed mainly in the Indo-Australa- sian Region, with a few described from the Palaearctic, Afrotropical and Oriental Regions (Bolton, 1995b). This ant distribution appears to be correlated with the number of mealybug species found in these regions. A few non-rhizoecine mealybugs may be associated with Acropyga in South America (Biinzli, 1935), and it seems unlikely that species of Acropyga in other re- gions will not exploit mealybug trophobionts in other subfamilies.
Biinzli (1935) has listed R. coffeae, R. caladii Green, R. moruliferus Green (= R. falcifer Kiinckel d’Herculais), Geococcus coffeae Green, Pseudo- rhizoecus proximus Green and Capitisetella migrans (Green) as attacking coffee roots in Brazil. All the indigenous host plants of these mealybugs are grasses, and Biinzli has shown that the mealy- bugs have migrated to the new host plant Coffea liberica.
Other rhizoecine mealybugs associated with Acropyga in SouthAmerica, attacking economic plants, are Neochavesia caldasiae (Balachowsky) on roots of coffee in Colombia, NV. eversi (Beardsley) on roots of banana in Panama, and N. trinidadensis (Beardsley) on roots of cacao in Trinidad.
In India, Dr C.A. Viraktamath, University of Agri- cultural Sciences, Bangalore, has sent specimens of X. acropygae sp. n., that were feeding on the rootlets of grape, causing a severe reduction in yield. The mealybug also attacked many weed species in the vicinity of the vines.
D.J. WILLIAMS
LIFE CYCLES
In the family Pseudococcidae, there are usually four instars in the female and five in the male. Normally the female is regarded as neotenic and all instars feed. In the male, feeding is restricted to the first two instars followed by the non-feeding prepupa, pupa and adult. Koteja (1985) suggests that loss of wings in the female occurred only once and was an adaptation to the original leaf-litter or hypogeic habitat. Normally adult males are much smaller than the adult female. Koteja (1985) hypothesizes that in the original leaf litter, small males were able to crawl among soil particles to reach the females. This small size could be realized only by a cessation of feeding, i.e. at the end of the second instar. In order for the males and females to emerge simultaneously, the last male instars became resting and non-feeding instars known as the prepupa and pupa.
The life cycles of Rhizoecus and near relatives are virtually unknown. Biinzli (1935) discusses four fe- male instars in R. coffeae and three male instars but he may have overlooked the pupal instars. Schmutterer (1952) discusses four female instars in R. albidus Goux in Germany and mentions nymphs and adults in the male without further details.
The adult males in some of the Rhizoecus group are wingless and morphologically degenerate but in other species they are winged (Schmutterer, 1952; Beardsley, 1962). Surprisingly, winged males in slide collections of Rhizoecus are not uncommon (D.R. Miller, per- sonal communication).
The first record of a female pupal instar was de- scribed by Williams (1988) for the genus Xenococcus. This instar replaces the normal third or feeding instar and probably acts as a resting stage for the female to develop in the presence of attending ants. A similar type of pupa has now been found and is discussed here in at least five species of Eumyrmococcus.
Unfortunately, no first-instar nymphs of Eumyrmo- coccus are known from any of the material at hand except for a single specimen still within the egg mem- brane. This first-instar nymph is densely covered in short setae and shows unusual developmemt of many characters as in the first instar of Xenococcus (Fig. 24). Second-instar nymphs are available in 11 species of Eumyrmococcus but these remain unsexed. Normally second-instar mealybugs show strong dimorphism, with the second-instar male usually possessing more tubular ducts and pores to form a waxy covering for the prepupa and pupa. These characters are usually absent or fewer in the second-instar female. In Eumyrmococcus, the female pupal instar and the male prepupal and pupal instars are without any coverings and any sexual differences in the second instar are not apparent.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 5
Female pupae show a well developed labium and this is evidenced by at least four species of female pupae with the developing adult female still inside (Fig. 1B). Itis thus easy to separate female pupae from male prepupae and pupae which lack the well devel- oped labium (Figs 21, 25). Male prepupae are available with the developing pupae still inside and pupae are at hand with the developing adult male inside just before the final moult (Fig. 1A). All female pupae, and male prepupae and pupae, lack dermal structures such as pores and setae. Unlike most male pupae in the Pseudococcidae, which usually possess longer limbs and antennae than those of the prepupae, these charac- ters in the pupa of Xenococcus are shorter, presumably in preparation for the adult male which possesses minute tubercle-like antennae and short legs.
Studies of the adult male of Rhizoecus are remark- ably few. Beardsley (1962) described the adult male of R. falcifer in detail and showed that the head bears only two pairs of eyes and lacks the normal lateral ocelli. Furthermore, the head is not separated from the thorax by a constricted neck. The penial sheath, ac- cording to Beardsley, is related to that of Phenacoccus Cockerell. Similar features are present in the male of R. albidus as shown by Schmutterer (1952).
The adult male of Neochavesia eversi (Beardsley) is morphologically degenerate, without wings, eyes or thoracic sclerotization, and the penial sheath and aedeagus resemble those of Puto (Beardsley, 1970). Adult males of three species of Eumyrmococcus are available, one of which is not described here because no adult females were collected. The males are dispa- rate. All are without eyes, wings and thoracic sclerotization and possibly resemble males of Neochavesia. The genital capsule in the Australian species studied, possesses a long dorsal style and a long pointed aedeagus. It is not clear why there should be such development when the tip of the female abdo- men is normally curled dorsally, presenting easy access. Adult males of E. sarawakensis sp. n. are elongate, curled ventrally, with most of the genital capsule internal. Both the male and female in any one species of Eumyrmococcus are about the same size.
Peculiar development in the adult male of Xenococcus is described on p. 24. The legs are placed at the anterior end of the body and the claws have the strangest development known in any scale insect, with apparent digitules larger and stouter than the actual claw.
PARASITISM
Of about 160 species described in the subfamily Rhizoecinae, only a single record of a parasitoid has been published. Schmutterer (1952) reported a very
effective chalcid parasitising R. albidus Goux in Ger- many. The parasitoid was recorded later as Anomalicornis tenuicornia Mercet (Hymenoptera: Encyrtidae) by Ferriere (1956), a fairly common Old World species.
The Encyrtidae have successively parasitised the other three mealybug subfamilies Trabutininae, Pseudococcinae and Sphaerococcinae recognized by Koteja (1974a, 1974b) as evidenced by the host parasitoid index of encyrtid parasitoids of mealybugs in Noyes and Hayat (1994). The views of Koteja (1984, 1985) that Rhizoecus, which we may substitute as the Rhizoecinae, could have spent the whole of their phylogeny underground, may be a reason for the lack of parasitism. Many of the mealybugs in the other three subfamilies live underground or are concealed and are known to be parasitised. The hypogeic habit is, therefore, not a barrier. Many Rhizoecinae have a symbiotic relationship with ants which could success- fully exclude parasitoids but conversely many Rhizoecinae are not associated with ants. Parasitoids may not recognize rhizoecine mealybugs as suitable hosts. J.W. Beardsley (personal communication) has commented that the Rhizoecinae may have branched off from the main mealybug lineage before the evolu- tion of the Encyrtidae and may not be recognized as normal encyrtid hosts. It is possible, however, that parasitoids have simply not been collected from the Rhizoecinae.
MORPHOLOGY OF EUMYRMOCOCCUS AND XENOCOCCUS (ADULT FEMALES)
Bopy. All descriptions of the adult female in life indicate that the cephalothorax is strongly dilated, or shows some sign of dilation, with a narrow tapering abdomen which curls dorsally. When prepared on microscope slides, the anal lobes are poorly developed and are recognizable by infolding on both the dorsum and venter with the apex of each lobe bearing three long setae, collectively referred to here as anal lobe setae. In E. neoguineensis sp. n., each lobe has 8 anal lobe setae and in E. /anuginosus sp. n., the anal lobe setae are not differentiated from other setae on the anal lobes. Usually there are 3 anal lobe setae that are long and stout at the bases with wide setal collars. They are similar to those in many other genera of the Rhizoecinae. In Eumyrmococcus and Xenococcus, there are usually one ventral and two dorsal setae on each anal lobe but occasionally two are placed ven- trally and one dorsally.
ANAL RING. The anal opening projects slightly be- tween the anal lobes and the anal ring is normally
6
simple with a crescentic sclerotized band without cells. Three pairs of anal ring setae appear to be usual but occasionally there are 4~7 pairs. The anal ring setae are normally detached from the anal ring, lying just anterior to it except in E. corinthiacus Williams, which possesses 4 pairs, apparently attached to the outer edge of the ring. In all species the anterior pair of setae are short and slender, the second pair either the same length as the anterior pair or longer, and the posterior pair often long and stout, sometimes as long as the anal lobe setae.
ANTENNAE. The antennae are 2—4-segmented and differ in shape and position. In E. lanuginosus, they are 2-segmented, short and tubercle-like. In most spe- cies of Eumyrmococcus the antennae are 2-segmented, with a short basal segment and a long tapering second segment. They are placed fairly wide apart on either the ventral or dorsal margin of the head. In E. sarawakensis sp. n. and E. sulawesicus sp. n., the antennae are situated well on the dorsum of the swol- len cephalothorax. The 4-segmented antennae of E. neoguineensis are unusual in possessing spine-like setae on the leading edge. Xenococcus has 4-seg- mented antennae that are long, sometimes as long as the body, with strong articulation between the first and second segments. The second segment is provided with small points at the proximal end which fit into grooves at the distal end of the first segment. This articulation allows the antennae to fold against the dorsum of the body. A pair of peg-like setae are present at the tip of each fourth segment.
LeGs. The legs are always well developed with few setae. The setae are stout and flagellate but sometimes a few on the outer edge of the femur in Eumyrmococcus are blunt and sensory. In Xenococcus, the leg setae may be unusually long on the distal end of the femur, almost as long as the femur. In most species the tibia is swollen distally before narrowing towards the tarsus. The tarsus is usually swollen near the base then tapers gradually, sometimes for nearly half its length. In one species of Xenococcus, the tibia and proximal end of the tarsus are straight. The claw in E. scorpioides (De Lotto) is broad but in all other species it is long, pointed and slender with a pair of short setose digitules. These digitules are difficult to discern in some species except for the bases. A campaniform sensillum is present at the base of all tarsi.
Eyes. These are completely absent in all species of Eumyrmococcus and Xenococcus.
LABIUM. Inmost species the labium is narrow with 3 segments, often longer than the clypeolabral shield. The setae on the anterior surface of the apical segment are often widely spaced and the subapical setae are sometimes reduced to 3 pairs. A few species possess 2 pairs of setae on the clypeolabral shield.
D.J. WILLIAMS
CEPHALIC PLATE. This sclerotized area on the frons in many species of Rhizoecus is not apparent in any species of Eumyrmococcus or Xenococcus.
CIRCULI. The type species of Eumyrmococcus, E. smithii Silvestri, is the only species studied without a circulus. In other species the circuli number 1-3, situated within the borders of the segments. When only a single circulus is present it is situated either on abdominal segment II or III. When 2 are present they lie on abdominal segments II and III, and a third in Eumyrmococcus is sometimes present on the metathorax. In Xenococcus, when a third is present, it lies on abdominal segment IV. Each circulus is round and may be small (the smallest studied only 8.75 um in diameter) or unusually large (up to 85.0 um in diam- eter). The outer edge of each circulus may be flat or raised slightly from the surrounding derm. In one species the circulus is almost conical but the centre in all circuli is cupped, sometimes deeply in Eumyrmococcus, or shallowly in Xenococcus, so that the circulus resembles a crater. A similar type of circulus is present in Neochavesia, and Beardsley (1970) has cautiously termed these ‘circulus-like pro- jections’. The form of the circulus in all these genera is different from that of other genera in the Rhizoecinae which possess circuli that are usually truncate-conical with the centre part flat and often reticulate or faveo- late.
A function of the circulus as an adhesive organ has been demonstrated for Pseudococcus adonidum (L.) (=P. longispinus (Targioni Tozzetti)) and Planococcus citri (Risso) by Pesson (1939) and in other species by Lloyd and Martini (1957). Ferris & Murdock (1936), from histological sections, discussed the circulus in Pseudococcus maritimus (Ehrhorn) as a secretory or- gan. Pesson (1939), in a detailed study, found that the circulus of Pseudococcus 1s an exsertile organ, lacking secretory pores between the organ and the cuticle and that the epithelium is only a part of the hypoderm. The epithelium is responsible for the formation of the peripheral chitin, which protects it, and of a mucin which hardens after secretion. In the circulus, the substance secreted appears to be a chitin which re- mains soft longer with the gluey properties of a mucin and hence gives the circulus adhesive properties. It is easy to see the exsertile action of the circulus, accord- ing to Pesson, by placing the mealybug on a microscope slide so that it has difficulty in taking hold with its claws. The mealybug can then hold on to the surface by the circulus.
It is not clear 1f the circulus of Eumyrmococcus and Xenococcus has the same function as those studied above. Silvestri (1924) studied histological sections of the circulus of X. annandalei and concluded that there are numerous elongate-pyriform cells directly under the shallow central cup. Silvestri thought that the
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI il
circulus secretes some kind of liquid which may be attractive to the ants attending the mealybugs. Any exsertile properties of the circulus in Eumyrmococcus and Xenococcus, or even in any Rhizoecinae, remain unproved. The term circulus is retained for the species under discussion.
OSTIOLES. These structures are always absent in Eumyrmococcus, Xenococcus and Neochavesia but are usually well defined in some other genera of the Rhizoecinae.
SETAE. One of the striking characters of Eumyr- mococcus and Xenococcus is the abundance of short setae on the dorsum of the abdomen. They extend to the venter of the abdomen in Eumyrmococcus and may be present on the head and thorax of both surfaces. In Xenococcus, they extend to the head and thorax of the dorsum and to the ventral lateral margins of the head and thorax. They lie in well defined bands across the segments leaving clear areas on the intersegmental lines. Normally these setae are flagellate, often slightly curved, but in E. falciculosus sp. n., they are mostly sickle-shaped, and in Xenococcus, there are patches of sickle-shaped setae among the flagellate setae on the dorsum of the thorax.
Stout and longer flagellate setae are usually present on the mid-venter of the thorax. In Xenococcus, these extend to the entire venter of the abdomen also. Sen- sory setae are sometimes present in Eumyrmococcus. These are slightly swollen at the apices and may be the only setae present on the head and thorax in some species. In other species, they are either long or short and sometimes mingle with the short, flagellate setae. The sensory setae may also be lanceolate or extremely minute and can be detected with certainty only by oil- immersion studies. Some short setae in E. neoguineensis are set in dermal pockets, forming clus- ters on the venter of the thorax.
MICROTRICHIA. Elongate microtrichia, resembling setae, are so abundant in some species that they almost completely cover the insect so that the setae are ob- scured. The setae in these species are only recognizable by the presence of setal collars. At present, species with dense microtrichia are only known from the more equatorial areas, Sumatra, Sarawak and Sulawesi.
PORES AND DUCTS. Some genera of Rhizoecinae contain species possessing trilocular pores, multilocu- lar disc pores and tubular ducts. Normally, trilocular pores secrete fine wax to protect the surface from contamination by the insect’s own honeydew. Ex- creted honeydew balls are often coated with this wax. Multilocular disc pores secrete wax to help in the production of ovisacs or to protect eggs. They are sometimes present in second-instar males to secrete wax in the formation of the cocoon (Cox and Pearce, 1983).
Tubular ducts in the Rhizoecinae are usually small, often narrower than the trilocular pores. Normally, tubular ducts in the adult female secrete wax to form the ovisac, or in the second-instar male, to help forma cocoon.
Pores and ducts are completely absent in Eumyrmococcus and Xenococcus. Honeydew excreted is probably taken entirely by the ants without need for the mealybugs to secrete wax as an anti-contaminent. Apparently the species are ovoviviparous and first- instar nymphs would be protected and transported by ants along with any other instars. Second instars lack pores and ducts and succeeding male prepupae and pupae are not enclosed in cocoons. The dense covering of short setae on the abdomen of all species and the complete covering of microtrichia in some species may help to protect the mealybugs from any possible contamination of honeydew and from excessive hu- midity in the nests and tunnels formed by the ants.
In Neochavesia, trilocular pores are found in the type species only but these pores are unlike any other trilocular pores in possessing a minute filament at the centre. Normal trilocular pores and tubular ducts are present in Leptorhizoecus gen. nov. (Fig. 28).
GENERA AND SPECIES
Eumyrmococcus and Xenococcus are closely related and it could be argued that Xenococcus is a subordi- nate clade withinEumyrmococcus. Both genera occupy similar geographical areas in southern Asia and Aus- tralasia (Figs 29, 30) and the species in both genera are mostly associated with specific ants of the same genus (p. 30). Xenococcus possesses unusual 4-segmented antennae, almost as long as the body, with remarkable articulationn between the first and second segments. In Eumyrmococcus, the antennae are much shorter and lack this special articulation. Furthermore, all species of Eumyrmococcus have a dense covering of short setae on both the dorsum and venter of the abdomen. In Xenococcus, short setae are restricted to the dorsum of the abdomen only and any ventral abdominal setae on the abdomen are long and stout. Although both genera have many shared characters, the genera are regarded here as distinct, in the interests of nomen- clatural stability, pending future cladistic studies. A key to separate both genera is included in the general key on p. 30.
Some species of Eumyrmococcus share a large type of circulus, others a small type, and two species lack any type. Although it may be possible to recognise species-groups based on the type of circulus, the circu- lus is not a stable character in some other genera of Rhizoecinae when any single species may possess 0— 2 (Williams, 1996).
Species-groups could be recognised on other char- acters. The E. smithii-group contains species with abundant short flagellate setae covering the dorsum and venter and lacking any sensory setae. In addition to the type species, E. smithii, this group also contains E. kolombangarae and E. kusiacus. Another group, found near the equator, comprises E. kruiensis, E. lanuginosus, E. maninjauensis, E. sarawakensis and E. sulawesicus. All are densely covered in elongate microtrichia. In possessing similar 4-segmented an- tennae, E. corinthiacus in Europe, is probably related to E. scorpioides from South Africa even though the claws in E. scorpioides are stout and different from the long slender claws in any other species. The Australian species, E. lamondicus, E. queenslandicus, E. recalvus and E. taylori appear to be related to the Japanese species, E. nipponensis. All possess similar 2-seg- mented antennae, with the second segment elongate and tapering, and with elongate posterior setae on the anal ring.
Most of the abundant abdominal setae in E. falciculosus are sickle-shaped, unlike the setae in any other species, although similar setae are present in small numbers in both species of Xenococcus. A final species, E. neoguineensis, has an unusual combination of characters. The 4-segmented antennae are the long- est of any species and, although the articulation between the first and second segments is normal, these anten- nae possibly come nearest in structure to those of Xenococcus. Furthermore, the anal lobes in E. neoguineensis contain numerous setae and the body shape is unusual, without such a marked dilated cephalothorax. This species may form a distinct group.
The two adult males described here in Eumyrmococcus are so disparate that if discovered without accompanying adult females, it would be difficult to place them in the same genus. The differ- ence between these adult males is as great as the difference between either and the adult male of Xenococcus acropygae. Similar differences in male morphology are expected among the different species- groups of Ewmyrmococcus discussed here.
Eumyrmococcus Silvestri Eumyrmococcus Silvestri, 1926: 271.
Type species Eumyrmococcus smithii Silvestri, by original designation.
DESCRIPTION
Adult female. Body pyriform, broadly pyriform, or elongate-pyriform, with cephalothorax usually strongly dilated, often constricted between second and third abdominal segments, then abdomen tapering to nar- row abdominal segment VIII. In life, tip of abdomen curled dorsally. Posterior end rounded, anal lobes not developed but recognizable by inner edges slightly
D.J. WILLIAMS
grooved, each lobe usually terminating in long, stout flagellate setae forming a group of 3, sometimes nearly as long as body, sometimes stout and short or in groups of numerous slender flagellate setae. Antennae each with 2-4 segments, placed fairly wide apart on venter of head margin or displaced to dorsum of head when prepared on microscope slides; terminal segment usu- ally tapering, sometimes tubercle-like or long and slender. Legs well developed, tibia and tarsus subequal in length, tarsus swollen from joint with tibia then tapering to narrow distal end, all segments with fairly stout flagellate setae, set well apart; occasionally with sensory setae on outer edges of femur and tibia. Claw normally long and slender, pointed, except in one species with claw relatively stout at base; each with a pair of short, slender setose digitules but these difficult to recognise in some species. Labium longer than wide, often with 3 pairs of subapical setae, widely spaced. Clypeolabral shield with 1 or 2 pairs of setae. Anal ring dorsal, sclerotized, usually crescentic except in one species when circular; simple, without cells; with normally 3 pairs of setae; first 2 pairs usually detached, lying just anterior to ring, second pair some- times longer than anterior pair, posterior pair sometimes stout and as long as anal lobe setae; sometimes ring with 4~7 pairs of setae, variously placed. Circuli present or absent; when present numbering 1-3, all round, small to unusually large and conspicuous, always deeply cupped.
Body setae short and abundant, at least on abdomen, sometimes present on entire body, usually flagellate but sometimes most abdominal setae replaced by sickle- shaped setae.
Setae on head and thorax often longer and stouter. Sensory setae, either lanceolate or with slightly swol- len blunt tip, sometimes distributed over entire body with the flagellate setae, or situated at anterior end of body only: occasionally minute bulbous setae present. Microtrichia sometimes present, covering entire body, these often nearly as long as setae giving the insect a woolly appearance and often masking the short body setae. Ostioles absent. Tubular ducts, multilocular disc pores and trilocular pores absent.
First instar. Well clothed in short setae as evi- denced by single specimen still within egg membrane.
Second instar. Similar to adult female but with fewer setae and usually shorter antennae and legs.
Female pupa. Recognizable by elongate labium, antennae showing segmentation, legs well developed.
Male prepupa and pupa. Similar to female pupa but with shorter limbs and antennae. Labium not de- veloped.
Adult male. Wingless and morphologically degen- erate, of diverse appearance. Genital capsule well developed, either exposed or mostly enclosed within abdomen; when exposed, with conspicuous anal open- ing and dorsal style, penial sheath rounded apically,
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 9
ventral slit rounded near base then widening apically, basal ridge of penial sheath well defined ventrally, aedeagus long and pointed. Posterior edge of abdomi- nal segment VIII forming possible anal ring as in female, with 3 pairs of setae. Legs well developed, slender. Body setae numerous. Antennae 2-segmented. Eyes absent.When genital capsule mostly enclosed within abdomen then ventral slit almost square, anal opening oval at posterior end of abdomen at base of sclerotized extension, possibly representing a short style. Legs squat. Antennae short, 2-segmented. Body almost naked except for a few minute setae.
COMMENTS.
The description of the adult male is
based on two species and another still within the pupal instar. The wide disparity in shape and form may indicate a much wider range of variation.
Key to Species of Eumyrmococcus (Adult
Females) PAIL AG SEPINIENLCH 5---.--.-..-.-ccorecceecceanssrargrececcerersers 2 SOILED UAG 2 -SEPIMENLEa..:<s<aceis.-a-bervsenequerexe<nskspuwonentebss 4 2 Anal lobe setae in groups of 3.Anal ring with 6-8 pairs of
QELS, = cotdcaned ree ROLE EPEC EEE ALE EERE EERE CELE Pen ret ae ar ee 3
Anal lobe setae in groups of 8 or 9. Anal ring with 10 pairs of setae (Fig. 10) ................ neoguineensis (p. 14)
All setae on head, thorax and abdomen, minute and flagellate. Claw elongate, slender (Fig. 2) .............::000+ stontan <ec SR RAR ORR Eee corinthiacus (p. 10)
Most setae on head and thorax sensory, blunt, expanded apically, only abdominal setae flagellate. Claw stout and ISHN (EM SO) ca an aaruasioxsaisse Pap ox saicvans al scorpioides (p. 19)
Surface of body densely covered in minute hair-like microtrichia in addition to abundant slender setae ...... 5
Surface of body not densely covered in minute hair-like microtrichia. Abundant setae present only, although there may be a few stout microtrichia present in some areas
Anal lobe setae not differentiated from other setae on anal lobes. Antennae each with segment 2 short, scarcely longer than segment | (Fig. 8)........ lanuginosus (p. 13)
Anal lobe setae in groups of 3 on each anal lobe. Anten- nae each with second segment noticeably longer than ISTE: CSL Ta Re ee enon Oe Re ae 6
Anal lobe setae and posterior anal ring setae short, about 70-75 \m long. All setae flagellate, blunt sensory setae SO tat (EU GD) (2 Fa teaches Sarsteeceeechcn se cess kruiensis (p. 11)
Anal lobe setae and posterior anal ring setae much longer, each 220-525 um long. Blunt sensory setae present, at least on head, thorax and outer edges of legs .............. y
Antennae distinctly placed dorsally on microscope prepara- tions. Blunt sensory setae present on dorsum of head and thorax in addition to short minute flagellate setae ...... .. 8
Antennae distinctly placed ventrally on microscope prepa-
9
10
rations. Blunt sensory setae absent from dorsum of head and thorax, only short minute flagellate setae present (Bigs 9) seca Reece maninjauensis (p. 14)
Long, blunt sensory setae, at least 17.5—22.5 um long, present on dorsum anterior to antennae only, near to head margin. Shorter blunt setae present on dorsum posterior to antennae on thorax and abdominal segment I (Fig. 14) Seda ge CEPA VFO CER UGS EC AEE EES sarawakensis (p. 18)
Long blunt sensory setae, at least 45—50 um long, more numerous on dorsum, present anterior and posterior to antennae. Shorter blunt setae absent from dorsum on thorax and abdominal segment I (Fig. 18).............008 Bs seed ac secrets nan diG ee eee ean Sines Si sulawesicus (p. 20)
Most setae on abdomen short, stout and sickle-shaped (Rie) z eevee AR: ee. dates falciculosus (p. 10)
Most setae on abdomen short and flagellate, sickle- SHAPE! SETAC: ADSCM LE) wcxccenccurescnddvcesscescsccxsconesicaseteaeores 10
Blunt sensory setae present, in addition to dense, short Mapellatesetaey 5 sete sees ens, eles) es 13
Blunt sensory setae absent, dense, short flagellate setae PRCSEDM OMY race ccare-tocde user ceatesanscerstnssehieveverscesesesteraccs I]
Circuli absent. Anal lobe setae and posterior anal ring setae long, at least S00 um long (Fig. 17) «0.0.0... eee BE Pe Ree eect eee eee ee smithii (p. 19)
Circuli present, small, on abdominal segments II and III. Anal lobe setae and posterior anal ring setae much shorter, AUIMOST GSMO O MMO Sieve. -t sceccecenenerreceieverecssnanaesees 12
Long, stout flagellate setae, at least 80-90 Lm long, present in rows at posterior edges of dorsal and ventral abdominal segments, these in addition to abundant minute flapellate Setae (E10. ©) ccn.-cscccaneeede caren kusiacus (p. 12)
Long, stout flagellate setae shorter, at most 35 um long, present medially on venter of abdomen only, in addition to abundant minute flagellate setae (Fig. 4) .......0....0. es SOE RS See? Se kolombangarae (p. 11)
Most sensory setae lanceolateniic.0:icreesvesesstecsess-e: 14 All sensory setae blunt and expanded apically .......... 15
A single circulus present only, on abdominal segment III, wide and conspicuous, at least 52-60 [tm wide. Slender, blunt sensory setae present on dorsum and venter of El oral ovaunt atalllis(ern ant syan iil Ul (PSS 102) Weeperceerereee earn cere cecocceecee: co eaoemntenpeerets penile et aster he queenslandicus (p. 16)
Two circuli present, these on abdominal segments II and III, small, at most 11.0-12.5 um wide. Slender, blunt sensory setae absent from abdominal segment II (Fig. 7) Be yest UU Ek Re ta te ch areca lamondicus (p. 12)
Short slender setae abundant on head and thorax. Obanal setae stout and long, about as long as anal lobe setae. A single circulus present, this on abdominal segment III (iiss Ul) at Bee fet or tees nipponensis (p. 15)
Short slender setae sparse on head and thorax. Obanal setae short and slender. Two circuli present, these on abdominal segments II and III ..........eeeeeeeeeeeeeeeeeees 16
10
16 Dorsal setae on head and thorax all sensory, of various Sizes) (ig sil) pee. Seen ees recalvus (p. 17)
— Dorsal setae on head and thorax mostly long and flagel- late, sensory setae few and minute (Fig. 19) ................. Stith 2 ee ee ee taylori (p. 20)
SPECIES DESCRIPTIONS
Eumyrmococcus corinthiacus Williams
(Fig. 2)
Eumyrmococcus corinthiacus Williams, 1993: 218. Holotype adult, Greece, Corinth (BMNH) [exam- ined].
DISTRIBUTION
Greece: Corinth, Perachora, near Lautraki, carried by swarming ant Acropyga sp., 7.x.1985 (A. Buschinger); without locality, carried by queens of Acropyga sp. during mating flight, x. 1992 (W.H.O. Dorow).
COMMENTS. This species was described recently by Williams (1993). Important characters are the 4-seg- mented antennae and thick setae in groups of 3 on the anal lobes, 2 in each group distinctly longer and stouter than the other. The dorsal and ventral setae are abundant, all flagellate, becoming less numerous on the thorax and head. As in other species, the circulus is concave and cup-shaped when viewed laterally.
The original material, the first known from Europe, was sent for identification by ProfessorA. Buschinger, Institut fiir Zoologie Technische Hochschule, Darmstadt, Germany. When discussing the habit, Buschingeretal., (1987) recorded the ant as Plagiolepis sp. and this name was used by Williams (1993). Pro- fessor Buschinger (personal communication) has since indicated that the ant is actually a species of Acropyga. The most likely species is A.(Rhizomyrma) paleartica Mennozzi, the only species known in Greece.
The accompanying figure first appeared in the Ento- mologist’s Gazette and is reproduced here from the original illustration with slight modification.
Eumyrmococcus falciculosus sp. n. (Fig. 3)
DESCRIPTION
Adult female on microscope slide membranous, elon- gate, largest specimen 1.15 mm long, 0.50 mm wide, widest at about mesothorax; head and thorax rounded, constricted slightly between abdominal segments II and III, abdomen tapering and narrowing between abdomi- nal segments VII and VIII, posterior end of body rounded, abdominal segment VIII 200 im wide at base.
D.J. WILLIAMS
Positions of each anal lobe with 2 stout dorsal setae, each about 188m long, and | ventral setaabout 125umlong, forming a group of 3. Antennae situated on ventral head margin, each 65-77 um long, with 2 segments; basal segment wide, second segment tapering. Legs well developed, slender; hind trochanter + femur 115— 118 um long, hind tibia+ tarsus about 100um long, claw slender, about 20 um long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.84—0.86. Ratio of lengths of hind tibia to tarsus 0.73—0.81.Tibiae swollen then tapering, tarsi widening near base then narrowly tapering. Leg setae pointed. Labium about 117—125 um long, as long as clypeolabral shield, 77-80 um wide; ratio of length to width 1.51—1.56. Circulus present, distorted in available specimens but at least 50m wide, situated near middle of abdominal segment III.Anal ring about 75 um wide, with 6 slender setae, the anterior pair each about 32.5 um long, the second and posterior pairs longer but not complete in available material.
Dorsal surface with unusual sickle-shaped setae, fairly crowded, present as far forward as abdominal segment III, those posteriorly on abdominal segment VIII each about 20 Um long, others on abdomen mostly 12.5 tm long. A few flagellate setae each 12.5 um long, present among the sickle-shaped setae. Setae on head, thorax and abdominal segments I and II, curved but fairly stiff and pointed, mostly about 15 um long, moderately numerous. Thicker sensory setae, each bluntly tipped and with large setal collars, few, distributed mainly around margins.
Ventral surface with a similar distribution of sickle- shaped setae, pointed setae and thick sensory setae, to those on dorsum. A few long, stout sensory setae present in medial area of thorax.
MATERIAL
HOLOTYPE. Adult 9, Indonesia: Sumatra, Krui, Kubu Prau, among primary forest litter, 800 m, 29.v.1991 (L. Deharveng & A. Bedos) (MNHN).
PARATYPE. Indonesia: Sumatra, same data as holotype. 1 adult 2 (BMNH).
Non-type material, Indonesia: Sumatra, Jambi Proy- ince, Rantau Pandan, in soil in Hevea plantation, 9.vi.1991 (L. Deharveng & A. Bedos).
COMMENTS. This is a curious species with peculiar sickle-shaped setae on the abdomen and short 2-seg- mented antennae. It has no close relatives thus far discussed here. The antennae are similar to those of E. lanuginosus sp. n., also described from Sumatra, but E. lanuginosus is extensively clothed in long microtrichia whereas inE. falciculosus any microtrichia are short and barely perceptible.
The specimen recorded from Rantau Pandan is extremely poor but is recognisable as this species by the sickle-shaped setae.
The epithet ‘falciculosus’ is based on the Latin word
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 11
‘falcicula’ meaning small sickle, and the Latin suffix ‘-osus’ meaning ‘abundance of’, referring to the num- erous setae of this shape.
Eumyrmococcus kolombangarae sp. n. (Fig. 4)
DESCRIPTION
Adult female on microscope slide broadly oval, about 0.90 mm long, 0.60 mm wide, widest at mesothorax, tapering to a constriction between abdominal seg- ments VII and VIII, base of abdominal segment VIII 180 tm wide. Posterior end of body rounded; posi- tions of each anal lobe with 2 ventral setae and | dorsal seta forming a group of 3, each seta about 80 Um long. Antennae lying on ventral head margin, each about 110 um long, with two segments, the second tapering. Legs well developed; hind trochanter + femur about 168 um long, hind tibia + tarsus 142 ttm long, claw slender, unusually long, about 45 um long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.84. Ratio of lengths of hind tibia to tarsus 1.84. Tarsi each swollen near base then tapering to narrow distal end. All leg setae stout and flagellate. Labium distorted in available specimen but at least 100 ttm long and fairly wide, about as long as clypeolabral shield. Circuli numbering 2, each about 15 um wide, situated near posterior edges of abdomi- nal segments II and III but within borders of segments. Anal ring about 75 lm wide, with 6 setae, each about 65 um long.
Dorsal surface with crowded flagellate setae occu- pying most of surface except for well marked intersegmental areas; setae on abdominal segment VIII mostly 15 {1m long, anteriorly about 12.5 um long; a few setae on any segment with larger setal collars.
Ventral surface with crowded setae as on dorsum except in medial area of thorax where they are longer and not numerous. Long stout setae, each about 35 um long, present in pairs near mid-line of abdominal segments II-VI.
MATERIAL
HoLoryPe. Adult ?, Solomon Islands: New Geor- gia Group, Kolombangara, E. Kusi, 600 m, associated with Acropyga lauta in log, 30.vili.1965 (Isiah) (ANIC).
PARATYPES. Solomon Islands: same data as holotype, 1 2nd instar (ANIC), 1 2 pupa (ANIC), | pupa (BMNH).
COMMENTS
This species is related to the type species E. smithii in the arrangement of the densely crowded flagellate setae, at least on the dorsum, but differs in possessing
two small circuli. E. smithii lacks circuli completely. Furthermore, in E. kolombangarae, the anal lobe setae are at most only 80 um long and the posterior anal ring setae are about 65 {1m long, whereas in E. smithii, these setae are noticeably much longer, at least SOO um long.
E. kolombangarae is very close to E. kusiacus, herein described, differing mainly in lacking long stout setae in rows at the posterior edges of the dorsal and ventral abdominal segments. InE. kolombangarae, long stout setae are present in pairs only near the midline of the venter of abdomen.
The description of this species has also been based on an almost perfect adult female still within the pupal instar. A single second instar is also available. This is similar to the adult female, differing mainly in having shorter limbs, a single circulus present on abdominal segment III, and in the body setae which are not so dense.
The epithet is based on the Latin genitive of the place name meaning ‘of’ or ‘from’.
Eumyrmococcus kruiensis sp. n. (Fig. 6)
DESCRIPTION
Adult female on microscope slide, elongate, membra- nous 0.8l1mm long, 0.38 mm wide, widest at mesothorax, abdomen gradually tapering, slightly con- stricted between abdominal segments VII and VIII, base of abdominal segment VIII 120 um wide, poste- rior end of body fairly straight. Positions of each anal lobe with 2 stout dorsal setae and | ventral seta, each about 70 um long, forming a group of three. Antennae placed on ventral head margin, each 82.5 um long, with 2 segments, the second segment 60 Lm long, tapering. Legs well developed; hind trochanter + fe- mur about 117.5 um long, hind tibia + tarsus 95 Um long, claw slender, about 27.5 um long, 5 um wide at base. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur ().80. Ratio of lengths of hind tibia to tarsus 1.11. Tibiae swollen then tapering to narrow distal ends. All legs with stout flagellate setae. Labium about 92.5um long, same length as clypeolabral shield, 65 um wide; ratio of length to width 1.42. Circulus large, about 45 ttm wide, situated near posterior end of abdominal segment III but within borders of segment. Anal ring about 70 um wide, with 6 setae; anterior and middle pairs slender, each 25—37 {1m long, posterior pair stout, each about 75 um long, similar to anal lobe setae.
Dorsal surface densely covered in hair-like microtrichia and slender flagellate setae except for intersegmental areas. Most setae on abdominal seg- ment VIII about 15 tm long, those on anterior abdominal segments 10.0—12.5 um long, most on head and thorax 5.0-—12.5 um long.
12
Ventral surface with similar microtrichia and setae to those on dorsum; on abdominal segment VIII, some setae as long as 40 um, most other setae on abdomen about 25 um long, marginal areas of head and thorax with short slender setae each about 12.5 tm long. Long, stout flagellate setae, each 25—30 Um long, with large setal collars, present on head and submarginal and medial areas of thorax.
MATERIAL
HOLOTYPE. Adult 9, Indonesia, Sumatra, Krui, Pahmungan, in soil of dammar plantation, 23.v.1991 (L. Deharveng & A. Bedos) (MNHN).
PARATYPES. Indonesia, Sumatra, same data as holotype, 3 2nd instars (MNHN), 2 2nd instars (BMNH).
COMMENTS. This is a distinctive species within the group extensively clothed in hair-like microtrichia. The stout setae in groups of three on the anal lobes and the posterior pair on the anal ring are much shorter than those in E. maninjauensis, E. sarawakensis and E. sulawesicus. E. kruiensis also differs from these three species in lacking any sensory setae with swollen tips.
There are also available five second instars. These are similar to the adult female in the dense covering of microtrichia but the limbs are shorter and the body setae are fewer.
The epithet is based on the place name ‘Krui’ and the Latin suffix ‘-ensis’ denoting origin.
Eumyrmococcus kusiacus sp. n. (Fig. 6)
DESCRIPTION
Adult female on microscope slide broadly oval, 1.12 mm long, 0.65 mm wide, widest at mesothorax, abdomen tapering, abdominal segment VIII 220 Um wide at base. Positions of each anal lobe with 2 stout dorsal setae and | ventral seta forming a group of 3, each about 100 um long. Antennae placed near ventral head margin, each 112—120 um long, with 2 segments, the second tapering, both segments with long stout setae. Legs well developed, slender; hind trochanter + femur about 200 um long, hind tibia + tarsus 158-162 uum long, claw slender, about 58 um long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.79-0.81. Ratio of lengths of hind tibia to tarsus 1.02—1.16. Tarsi each swollen near base then tapering to narrow distal end. All legs with stout flagellate setae. Labium 150—162tm long, about same length as clypeolabral shield, 75—80 [1m wide; ratio of length to width 1.50-1.62. Circuli numbering 2 situ- ated near posterior edges of abdominal segments II and III but within borders of segments, diameter on abdominal segment II about 10 Lm, on abdominal
D.J. WILLIAMS
segment III about 17.5 um, each deeply cupped. Anal ring about 100 um wide with 6 setae; anterior and second pairs each 75 Um long, posterior pair each about 87 Lum long, the anterior pair with setal collars touching.
Dorsal surface with crowded flagellate setae; on abdominal segment VIII mostly about 40 um long, anteriorly on abdomen and remainder of body mostly about 15 um long except towards posterior edges of abdominal segments, mostly about 25 um long. Setae on abdomen in fairly narrow bands leaving extensive intersegmental bare areas. Setae on head and thorax much more crowded than on abdomen. Long stout setae, each 80—90 um long, present in rows at posterior edges of abdominal segments II—-VII.
Ventral surface with similar setae to those on dor- sum, sparse on venter of thorax. Long stout setae as on dorsum, present at posterior edges of abdominal seg- ments I—VII.
MATERIAL
HoLotyPe. Adult ?, Solomon Islands: New Geor- gia Group, Kolombangara, Kusi, 1220 m, associated with Acropyga lauta, in log, 29.vii.1965 (Isiah) (ANIC).
PARATYPE. Solomon Islands: same data as holotype, 1 adult 2? (BMNH).
COMMENTS. E.kusiacus is related to E. smithii but differs in possessing two small circuli, and short stout setae on the anal lobes. These setae are scarcely longer than the width of the anal ring. In E. smithii, the circuli are absent and the setae on the anal lobes are conspicu- ously longer than the width of the anal ring. E. kusiacus is very close to E. kolombangarae, differing mainly in possessing long stout setae on the dorsum and venter at the posterior edges of the abdominal segments. In E. kolombangarae, there are long stout setae in the me- dial areas of the ventral abdominal segments only.
The name is based on ‘Kusi’, the place of origin, and the Latin suffix “-acus’ meaning “belonging to’.
Eurmyrmococcus lamondicus sp. n. (Fig. 7)
DESCRIPTION
Adult female on microscope slide elongate, slender, largest specimen 1.28 mm long, 0.53 mm wide (holotype 1.40 mm long, 0.40 mm wide), widest at about mesothorax, head and thorax rounded, narrow- ing at abdominal segment III then narrowly tapering to small constriction between abdominal segments VII and VIII, base of abdominal segment VIII about 100 um wide. Positions of each anal lobe with 2 stout dorsal setae and | ventral seta, each about 325 um long, forming a group of 3. Antennae placed on dorsal head margin, each 330-340 um long, with 2 segments, the
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 13
second 260-280 um long, tapering. Legs well devel- oped, slender; hind trochanter + femur 193-218 um long, hind tibia + tarsus 112-122 um long, claw slender, about 27.50 um long, 6.25 um wide at base. Ratio of lengths of hind tibia + tarsus to hind tro- chanter + femur 0.91—1.00. Ratio of lengths of hind tibia to tarsus 1.38—1.43. Tibiae swollen then tapering abruptly to narrow distal ends. All legs with stout pointed setae. Labium 127.5—137.5 tm long, slightly longer than clypeolabral shield, 80-105 um wide; ratio of length to width |.21—1.66. Circuli numbering 2, situated near middle of abdominal segments I and II, cup-shaped, about as deep as wide; on abdominal segment II about 12.5 tum wide, on abdominal seg- ment III about 11.0 um wide. Anal ring about 45 um wide, with 6 setae; anterior pair slender, each about 72 um long, second and posterior pairs stouter, each about 187 tm long, resembling anal lobe setae.
Dorsal surface with short crowded setae in bands across segments as far forward as prothorax; setae at posterior end of abdominal segment VIII each about 30 um long, most on abdominal segment VII about 12.5 um long, anteriorly mostly 7.5—10.0 um long; a few on most segments with larger setal collars; on prothorax and mesothorax setae occupying medial areas except for marginal band, absent from head and in noticeable intersegmental areas. Stout flagellate setae, 32.5-37.5 tum long, present in medial area of head, interspersed with a few smaller setae. Short lanceolate sensory setae, each about 10 Um long, not numerous, mingled with the slender setae posteriorly to abdominal segment VI. Slender sensory setae, each about 10 um long, and bluntly tipped, represented by a few on abdominal segment II only, these in addition to lanceolate setae.
Ventral surface with similar slender setae to those on dorsum, those on venter of thorax in deep marginal bands. Lanceolate sensory setae as on dorsum, few, present on most segments. Slender sensory setae oc- curring on abdominal segment II only. Stout flagellate setae present on head and medial area of thorax.
MATERIAL
HoLotyPe. Adult 2, Australia: Queensland, Mt Lamond, 12°44’S 143°18’E, associated with Acropyga sp., in rotting wood, 9—15.vi.1971 (R.W. Taylor & G.J. Feehan) (ANIC).
PARATYPES. Australia: Queensland, same data as holotype, 3 adult9? (ANIC), 2 adult 9? (BMNH), 4 2nd instars (ANIC), 3 2nd instars (BMNH).
COMMENTS. The distribution of the dorsal and ven- tral slender setae and the shape of the antennae, place this species near E. nipponensis. E. lamondicus, how- ever, comes closest to E. queenslandicus sp. n. in possessing minute lanceolate sensory setae, whereas those of E. nipponensis are blunt and swollen apically.
E. lamondicus differs from E. queenslandicus in pos- sessing also some slender blunt sensory setae on abdominal segment II and having two small circuli. In E. queenslandicus the blunt sensory setae are absent and there is only a single but large circulus. The slender setae on the venter of the thorax in E. lamondicus are also in much deeper marginal bands than in E. queenslandicus.
The epithet ‘/amondicus’ is based on the place name Mt Lamond and the Latin suffix *-icus’ meaning ‘be- longing to’.
Eumyrmococcus lanuginosus sp. n. (Fig. 8)
DESCRIPTION
Adult female on microscope slide elongate-oval, mem- branous, largest specimen 1.00 mm long, 0.50 mm wide, widest at mesothorax then gently tapering to abdominal segment VII, with a constriction between abdominal segments VII and VIII, posterior end of body rounded. Positions of each anal lobe with numer- ous long stout setae, mostly about 75 [Um long. Antennae placed on ventral margin of head, short, each 35,0-37.5 tum long, with 2 segments, second segment tapering, almost conical. Legs well devel- oped, robust, hind trochanter + femur 145-150 um long, hind tibia + tarsus 97.5-110.0 um long, claw slender, 10.0-12.5 tum long, 7.5 tum wide at base. Ratio of lengths of hind tibia + tarsus to hind tro- chanter + femur 0.67—0.73. Ratio of lengths of hind tibia to tarsus 1.00-1.09. Tibiae each swollen near base then tapering to narrow distal end. All legs with stout flagellate setae. Labium 95—100 um long, 50.0— 62.5 um wide, ratio of length to width 1.6—1.9. Circuli numbering 2 or 3, when 3 then situated near middle of mesothorax and abdominal segments II and II]. When only 2 then absent from mesothorax; each circulus about 12.5 um in diameter except on abdominal seg- ment III when about 10 Um in diameter, truncate-conical but deeply cupped from apex. Anal ring 70—74 um wide, with 6 setae; anterior pair slen- der, each about 30 um long, second pair each about 50 um long, posterior pair stouter, each about 80 um long; anal ring setae difficult to discern because of surrounding setae.
Dorsal surface extensively covered in fine hair-like microtrichia and short fine setae except for bare in- tersegmental areas. Short setae on abdominal segment VII mostly about 15 um long, those anteriorly mostly about 7.5 um long except for a few about 12.5 um long with slightly larger setal collars. Long stout setae present towards posterior edges of abdominal seg- ments IV—VUI, those on abdominal segment VIII numerous, 55—75 1m long; anteriorly mostly 55 um long, distributed across the segments except on ab- dominal segment IV where they are sparse.
14
Ventral surface with similar microtrichia and setae to those on dorsum, covering most of surface; the short setae fine in medial area of thorax. Long flagellate setae, much more numerous than on dorsum, present towards posterior edges of abdominal segments and in medial areas of head and thorax. A few stout sensory setae, each with swollen tip, present laterally on thorax.
MATERIAL
HoLotyPe. Adult ?, Indonesia: Sumatra, Si Antar [2°40’N 98°43’E], with ants (Smithsonian Expedi- tion), 1937(W.M. Mann) (USNM).The holotype is the middle specimen of five, all mounted in a row on the same slide and is clearly marked.
PARATYPES. Indonesia: Sumatra, same data as holotype, 3 adult 9, clearly marked in row with holotype, 1 2nd instar to right of other specimens (USNM).
COMMENTS. E. lanuginosus belongs to the group of species extensively covered in minute hair-like microtrichia. It differs from all the other species in possessing only short, almost conical antennae, and numerous stout setae on the positions of each anal lobe, not differentiated into groups of three.
The specimens were first studied by Harold Morrison who noted on the envelope that the species had been discussed by Roepke (1930). Although Roepke’s illus- tration of the mealybug is rather simple, it does show the small tubercle-like antennae. The ant species was provisionally identified as Cladomyrma sp. but the equally simple illustration of the ant with the mealybug held in the mandibles, may be a species of Acropyga.
The epithet ‘lanuginosus’ is a Latin adjective mean- ing downy or woolly, referring to the dense covering of microtrichia.
Eumyrmococcus maninjauensis sp. n. (Fig. 9)
DESCRIPTION
Adult female on microscope slide, membranous, pyri- form, about 0.83 mm long, 0.48 mm wide, widest at mesothorax, narrowing abruptly at about abdominal segment II, then gently tapering to rounded posterior end; abdominal segment VIII 150 tm wide at base. Positions of each anal lobe with 2 stout dorsal setae and | ventral seta, each about 275 um long, forming a group of 3. Antennae situated on ventral head margin, each 95 im long, with 2 segments, the second segment about 62.5 um long, tapering, | antenna partly divided on | side only. Legs well developed, slender, hind trochanter + femur about 107.5 tum long, hind tibia + tarsus about 105 tm long, claw, slender, about 25 um long, 5.0 um wide at base. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.97. Ratio of lengths of hind tibia to tarsus 1.0. Tibiae each narrow
D.J. WILLIAMS
at base, widening, then tapering abruptly to narrow distal end. All legs with outer setae sensory, each with slightly swollen tip, inner setae flagellate. Labium fairly wide, difficult to measure in available specimen, about same length as clypeolabral shield. Circulus large, about 32.5 lum wide, placed near middle of abdominal segment III. Anal ring about 75 um wide, with 6 setae; anterior pair slender, each 67.5 tum long, second pair similar, each about 70 um long, posterior pair stout and long, about 220 um long, resembling anal lobe setae.
Dorsal surface densely covered in hair-like microtrichia and short, fine setae, except for interseg- mental areas. Setae on abdominal segment VIII flagellate, mostly about 20 tm long; similar setae 10— 20 um long, present towards posterior edges of abdominal segments [V—VII; anteriorly, setae much more slender, 6.25—12.00 long, sometimes difficult to discern amongst the dense microtrichia.
Ventral surface with similar microtrichia and setae to those on dorsum. Obanal and cisanal setae present, each about 50 um long; stout flagellate setae present on abdominal segment VIII, as long as 67 um but most at posterior edge of abdominal segments about 50 um long, mixed with others about 20 um long. Minute slender setae distributed around margins of thorax. Medial area of thorax with stout flagellate setae; me- dial area of head and lateral area of thorax with a few stout sensory setae, each 15—20 um long, with slightly expanded tips and with wide thick setal collars.
MATERIAL
HOLoTyPe. Adult 9, Indonesia: Sumatra, Barat Province, Kotomalintang, nr Lake Maninjau, in soil, extracted by Berlese funnel apparatus, 19.xii.1994 (L. Deharveng & A. Bedos) (MNHN).
COMMENTS. Among the group of species with a dense covering of hair-like microtrichia, E. maninjauensis seems to be related to E. sarawakensis and E. sulawesicus in possessing long stout setae in groups of three on the positions of each anal lobe, and a pair of similar posterior setae on the anal ring. The antennae of E. sarawakensis and E. sulawesicus, how- ever, are placed well on the dorsum of the cephalothorax, whereas in E. maninjauensis they are placed on the ventral head margin. Furthermore, any stout sensory setae with swollen tips in E. maninjauensis are distributed on the venter only, whereas in the other two species they occur on the dorsum also.
The epithet is based on the place name ‘Lake Maninjaw’ and the Latin suffix ‘-ensis’ denoting local-
ity. Eumyrmococcus neoguineensis sp. n. (Fig. 10)
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 15
DESCRIPTION
Adult female on microscope slide membranous except for abdominal segment VIII lightly sclerotized; elon- gate-oval, largest specimen 1.15 mm long, 0.57 mm wide, widest at about abdominal segments II and III; head and thorax rounded, anterior margin of head sometimes straight, constricted slightly at mesothorax, then widening to anterior abdominal segments, re- mainder of abdomen tapering, base of abdominal segment VIII about 230 ttm wide, apparent anal lobes set well apart, apex of abdomen protruding. Position of each anal lobe with 3 or 4 dorsal setae and usually 5 ventral setae forming a group of 8 or 9; inner ventral setae slender, each about 150 Um long, others increas- ing in length distally, outer ventral and dorsal setae stout, 550-600 tum long. Antennae situated on dorsal head margin, each 420-440 um long, with 4 segments, second segment longest with a row of short conical setae on anterior edge, increasing in length distally; setae on third and fourth segments long, longest on third segment, about 325 lum long. Legs well devel- oped; hind trochanter + femur 188—232 um long, hind tibia + tarsus 180-220 um long, claw slender, about 34 um long, 7.5m wide. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.93—0.95. Ratio of lengths of hind tibia to tarsus 1.20—1.29. First legs with noticeably larger coxae than in other legs and longer trochanter + femur, 232—245 um long, and longer tibia + tarsus, 260-272 um long. Tibiae of all legs swollen near bases and narrowly tapering distally. All legs with long slender setae, except for some stout setae on posterior surface of femur of each first leg. Labium 162-175 tm long, 87-100 tum wide, longer than clypeolabral shield; ratio of length to width 1.62—1.86. Circulus present near anterior end of abdominal seg- ment II but placed well within borders of segment, 20-25 um in diameter, deeply cupped, about as deep as wide or slightly deeper. Anal ring about 90 um wide, with 14 slender setae, each about 110 Lm long.
Dorsal surface densely covered in slender flagellate setae except around antennal bases and intersegmental areas; some setae at posterior end of abdominal seg- ment VIII each about 87 um long, those at anterior end of segment about 25 um long. Anteriorly on abdomen, as far forward as abdominal segment V, mostly 17— 25 um long; in medial area of abdominal segment VI, medial and anterior areas of abdominal segment V, and all anterior segments of abdomen, thorax and head, extensively covered in minute setae 10.0-12.5 um long. Posterior edges of abdominal segments each with a row of stout setae, each about 100 [1m long, a few at lateral edges of abdominal segments each about 125 um long.
Ventral surface with a similar distribution of setae to that on dorsum but mainly bare on medial areas of head and thorax except for a few short setae. A bunch of invaginated setae present lateral to each first and
second coxa, these lightly sclerotized when prepared on slides.
MATERIAL
HoLotyPe. Adult?, Papua New Guinea, East Sepik Province, Yawasora [ Yauwasoru], near Wewak, asso- ciated with Acropyga (Atopodon) ambigua Emery, in rotting log, 4—6.vu1.1972 (R.W. Taylor) (ANIC).
PARATYPES. Papua New Guinea, same data as holotype, 3 adult 9? (ANIC), 2 adult 9? (BMNH); same data but in mountain log, 1 adult ? (BMNH); Morobe Province, 16 km NW Lae, “Timber Track’, in rotting wood, with A. (Atopodon) ambigua, 12.vi.1976 (R. W. Taylor), \ adult 9 (ANIC).
COMMENTS. This species shares withE. corinthiacus and E. scorpioides the character of 4-segmented an- tennae but the antennae of E. neoguineensis are much longer, with long setae on the third and fourth seg- ments, and conical setae on the anterior edge of the second segment. Furthermore, the long stout setae on each anal lobe number eight or nine, whereas those in the other two species are in groups of three. The affinities of E. neoguineensis are obscure but Xenococcus annandalei also possesses large groups of long stout setae on the anal lobes and the antennae are also 4-segmented with long setae on the third and fourth segments. In Xenococcus, however, the anten- nae are much longer, about as long as the body, and there is well developed articulation between the first and second segments. The bunches of invaginated setae in E. neoguineensis have not been seen in any other species. Despite the unusual characters, the true affinities of this spectacular species appear to be with the genus Eumyrmococcus to which it is here assigned.
The epithet is a Latinized word for New Guinea with the Latin suffix ‘-ensis’ meaning ‘belonging or pertaining to’.
Eumyrmococcus nipponensis Terayama
(Fig. 11)
Eumyrmococcus nipponensis Terayama, 1986: 509; 1988: 645; Ben-Dov, 1994: 152. HOLOTYPE 9°, Japan, Noboritachi, Mikura-jima Is, Tokyo, on rootlets of plants in nests of Acropyga [Atopodon] nipponensis Terayama, 30.vi.1980 (K. Masuko) (NIAES).
DESCRIPTION
Adult female on microscope slide pyriform, largest specimen 1.40 mm long, 0.60 mm wide, widest at mesothorax, abdomen tapering, constricted slightly between abdominal segments VII and VIII, base of abdominal segment VIII about 180 um wide. Position of each anal lobe with 2 long, stout ventral setae and | dorsal seta, forming a group of 3, each seta about
16
450 um long. Antennae situated on ventral head mar- gin, each 2-segmented, 260-270 \1m long, second segment tapering, 210-215 um long. Legs well devel- oped, slender; hind trochanter + femur 180-200 um long, hind tibia + tarsus 182-195 tm long, claw slender, about 30 um long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.95-1.01. Ratio of lengths of hind tibia to tarsus 1.05—1.21. Tibiae swollen near middle then narrowly tapering. All legs with outer setae stout, sensory, swollen apically; inner setae flagellate. Labium 138—150um long, 75.0- 77.5um wide, slightly longer than clypeolabral shield; ratio of length to width 1.84—1.93. Circulus present, round to slightly elliptical, 20-30 um wide, deeply cupped, about as deep as wide, situated near middle of abdominal segment II. Anal ring 68-75 tm wide, with 6 setae; anterior pair slender, each about 27 um long, second pair stouter, each about 62 [im long, posterior pair stout and long, each about 450 um long, resembling anal lobe setae.
Dorsal surface extensively covered with short slen- der flagellate setae, sparse near head margin and absent in wide intersegmental areas; at posterior end of body each seta about 15 um long, anteriorly on abdominal segment VIII mostly 12.5 um long, elsewhere mostly about 7.5 tum long. Short stout sensory setae, each with noticeable swollen tip, fairly numerous across segments and mixed with the short flagellate setae; at posterior end of body, each about 11 um long but most anteriorly each about 7.5 um long. Head with longer stout sensory setae 12.5—25.0 um long. Most setae with setal collars probably slightly raised from sur- rounding derm so that setae appear to be surrounded by halos.
Ventral surface with similar short setae as on dor- sum but not so dense; absent from medial area of thorax. Long sensory setae, each about 35 [1m long, present at posterior edge of abdominal segment VII; similar setae, each about 25 tim long, present on head. Other sensory setae shorter, those around margins, each 7.5m long as on dorsum, others 12—28 um long, distributed across segments. Long, stout flagellate setae present in medial area of thorax. Three very slender sensory setae sometimes present on each side of circulus. Obanal setae stout and long, about as long as anal lobe setae. Many flagellate setae at posterior end of body with thick setal collars.
MATERIAL
Japan: Yakushima Is, Amboh, on rootlets of plants in nests of Acropyga (Atopodon) nipponensis Terayama, 17.x.1984(M. Terayama); Tokunoshima Is, Tampatsu- yama, same data but 8. viii. 1984 (all labelled paratypes).
COMMENTS. E. nipponensis differs from E. smithii, the only other species known from Japan, in possess- ing a circulus and sensory setae with expanded tips. In E. smithii, the circulus is absent and all setae are
D.J. WILLIAMS
flagellate. E. nipponensis seems to have affinities with two Australian species, E. lamondicus and E. recalvus, in possessing similar 2-segmented antennae, each with the second segment long and tapering; also with minute flagellate setae on the dorsum distributed at least as far forward as the prothorax. Most of the sensory setae in the two Australian species, however, are lanceolate, whereas those in E. nipponensis are blunt and widely expanded distally.
The ant attending this mealybug was described originally by Terayama (1985), from material col- lected at the same time.
The accompanying illustration has been prepared from paratypes kindly made available by Dr M. Terayama.
Eumyrmococcus queenslandicus sp. n. (Fig. 12)
DESCRIPTION
Adult female on microscope slide elongate, largest specimen 1.30 long, 0.65 mm wide, widest at mesothorax, head and thorax rounded, abdomen taper- ing to rounded posterior end, constricted between abdominal segments VII and VIII, segment VIII about 150 um wide at base; positions of each anal lobe with 2 dorsal setae and | ventral seta, each about 300 um long, forming a group of 3. Antennae situated on dorsal head margin, each 270-300 um long, with 2 segments, the second tapering, 220-240 tm long. Legs well developed, slender; hind trochanter + femur 190-192 tm long, hind tibia + tarsus 110-117 um long, claw slender, about 27.5 um long, 5.0m wide at base. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.97—1.02. Ratio of lengths of hind tibia to tarsus 1.37—1.41. Tibiae swollen then tapering towards narrow distal ends. All legs with stout pointed setae. Labium 130-138 um long, slightly longer than clypeolabral shield, 75—100 um wide; ratio of length to width 1.38-1.70. Circulus present near middle of abdominal segment II, large and conspicuous, 52— 60 tm in diameter, deeply cupped with almost parallel sides but wider than deep. Anal ring 52.5-60.0 Um wide, with 6 setae; anterior pair slender, each about 100 um long, second pair thicker, each about 220 um long, posterior pair each about 250 Um long, almost as thick and as long as anal lobe setae.
Dorsal surface with crowded slender flagellate se- tae as far forward as thorax, those on prothorax represented by a medial band, absent laterally on mesothorax leaving bare areas on head and laterally on prothorax, mesothorax and intersegmentally. Setae towards posterior end of abdominal segment VIII each about 37.5 um long, anteriorly on next few segments mostly 10-20 tm long, on thorax and anterior ab- dominal segments mostly 10 Um long; some setae on any segment with larger setal collars than others. A
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 17
few sensory lanceolate setae, each about 10 [1m long, present on thorax and abdominal segments I—V. Setae on head sparse, mostly stout and flagellate, 35-42 um long.
Ventral surface with similar setae to those on dor- sum. Short slender setae present around margins only of thorax and first abdominal segment, few. Long stout setae, 35—42 um long, not numerous, present on head and medial area of thorax. Sensory lanceolate setae, same as on dorsum, few, distributed on abdominal segments II—V.
MATERIAL
HoLoryePe. Adult 9, Australia: Queensland, Kir- rama Range, NE slope of Mt Pershouse, 700 m, associated with Acropyga sp., 5.v.1969 (E. Kennedy & R.W. Taylor) (ANIC).
PARATYPES. Australia: Queensland, same data as holotype, 7 adult 92 (ANIC), 5 adult 9? (BMNH), 1 adult? (USNM), | 2nd instar (ANIC), | 2nd instar (BMNH); 4.5 km W of Cape Tribulation, 760 m, taken in the mandibles of Acropyga sp. in flight, 20— 23.iv.1983 G.B. Monteith & D.K. Yeates), | adult 2 (QM).
COMMENTS. Among the species already described, E. queenslandicus comes nearest to E. nipponensis in possessing similar antennae, each with a long second segment, and with the thorax and abdomen densely covered in slender setae. In E. nipponensis, the sen- sory setae are blunt and expanded apically, whereas in E. queenslandicus they are lanceolate. E. queenslandicus comes closest to E. lamondicus sp. n. Both species have short lanceolate sensory setae but in E. queenslandicus there is a large single circulus only, present on abdominal segment III. In E£. lamondicus there are two small circuli present on each of abdomi- nal segments II and III. Furthermore, E. lamondicus possesses some short, blunt sensory setae on abdomi- nal segment II in addition to lanceolate sensory setae. These blunt setae are absent from E. queenslandicus.
Two second instars are also available. These have a similar appearance to the adult females but the short flagellate setae are present only as far forward as abdominal segment IV.
The epithet is based on the locality ‘Queensland’ and the Latin suffix *-icuws’ meaning ‘belonging to’.
Eumyrmococcus recalvus sp. n. (Fig. 13)
DESCRIPTION
Adult female on microscope slide, narrow, elongate, 1.40 mm long, 0.65 mm wide, widest at about metathorax; head and thorax rounded, constricted slightly between abdominal segments III and IV, then tapering to posterior end of body, sharply narrowing
between abdominal segments VII and VIII; abdominal segment VIII 162 tum wide at base before tapering abruptly, projecting beyond positions of anal lobes. Each anal lobe with a group of 3 long stout setae, 2 on dorsum and | on venter, each seta at least 240 um long but difficult to measure in available specimen. Anten- nae each about 270 um long, with 2 segments, the second segment curved, 205 tm long. Legs well de- veloped, slender; hind trochanter + femur about 170 um long, hind tibia + tarsus about 140 tm long, claw slender, about 28 um long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.82. Ratio of lengths of hind tibia to tarsus 1.8. Inner leg setae flagellate, outer setae sensory, with slightly swollen tips. Tibiae swollen towards distal ends then narrow- ing, tarsi expanding then tapering to narrow distal ends, no thicker than bases of claws. Labium about 112 um long, 82.5 im wide, about same length as clypeolabral shield; ratio of length to width 1.35; all labial setae flagellate. Circuli numbering 2, each about 10 um wide, situated near middle of each of abdomi- nal segments II and III, deeply cupped. Anal ring about 48 um wide with 6 setae; 2 anterior pairs slender, curled in available specimen but about twice length of diameter of anal ring, posterior pair thicker, almost as long as apical setae.
Dorsal surface with crowded slender setae as far forward as abdominal segment II; those on abdominal segment VIII each about 25 tm long, on abdominal segments V—-VII mostly 17.5 um long, anteriorly about 12.5 um long; some on any segment with larger setal collars. Setae on head, thorax and abdominal segment I, all sensory with only minutely swollen tips, not numerous; many on head and medially on thorax each up to 37.5 um long: others on thorax and abdominal segment I 12.5 um long, mingled with minute and slightly swollen setae each about 3.7 Lim long.
Ventral surface with similar setae to those on dor- sum as far forward as abdominal segment II; anteriorly not numerous; metathorax and abdominal segment I with short sensory setae as on dorsum, on margins of head and thorax mostly long and sensory, up to 37.5 um long; others on medial areas of thorax, long and flag- ellate.
MATERIAL
HoLoryePe. Adult 2, Australia, Queensland, Byfield, 22°51’S, 150°39’E, associated with Acropyga sp., 26.x.1976 (R.W. Taylor, T-A. Weir) (ANIC).
PARATYPES. Australia, same data as holotype, 4 2nd instars (ANIC), 2 2nd instars (BMNH).
COMMENTS. This species is very close to E. taylori sp. n., differing mainly in the shape of the setae anterior to abdominal segment II. In E. reclavus, these setae on the dorsum and around the ventral margins are all sensory and slightly expanded apically, the only
18
flagellate setae are those situated medially on the venter of the thorax. In E. taylori, all setae in these positions are flagellate.
Material at hand also contains six second instars. These resemble the adult female but the minute flagel- late setae are not so crowded and the appendages are slightly smaller.
The epithet ‘recalvus’ is the Latin adjective mean- ing “bald in front’, referring to the paucity of setae at the anterior end of the body.
Eumyrmococcus sarawakensis sp. n. (Figs. 14, 15)
DESCRIPTION
Adult female on microscope slide extremely slender, largest specimen 1.35 mm long, 0.52 mm wide, widest at about prothorax, then gently tapering, abdominal segment VII about 115 um wide at base. Position of each anal lobe with | stout dorsal seta and 2 ventral setae forming a group of 3, each 430-488 um long. Antennae placed well on dorsum of cephalothorax, each 260-267 tm long, with 2 segments, second tapering, 210-215 tm long. Legs well developed, slender; hind trochanter + femur 162.5—182.0um long, hind tibia + tarsus 160-185 um long, claw slender, about 22.5 um long, 6.25 lum wide at base. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.97-1.01. Ratio of lengths of hind tibia to tarsus 1.00— 1.05. Outer setae on legs sensory, with swollen tips, inner setae flagellate. Labium about 137 tm long, longer than clypeolabral shield, 70.0-82.5 um wide, ratio of length to width 1.66—1.96. Circulus large, round to slightly elliptical, 75—85 um wide, situated in middle of abdominal segment III, cupped or bowl- shaped, not so deep as wide. Anal ring 65-70 Um wide, with 6 setae; anterior pair each about 32.5 um long, second pair each about 62.5 tm long, posterior pair about 390 um long, similar to anal lobe setae.
Dorsal surface with minute setae and hair-like microtrichia covering entire surface except for in- tersegmental areas. Setae on abdominal segment VIII each about 10.0—12.5m long, others anteriorly mostly about 5 wm long, difficult to discern among microtrichia. Stout sensory setae with swollen tips, present on head, each 17.5— 22.5 um long, not numer- ous. A few sensory setae, each about 12.5 um long, present on thorax and abdominal segment I.
Ventral surface with similar setae and microtrichia to those on dorsum. Stout sensory setae present on head and thorax, not numerous; short sensory setae, 10-15 um long, few, distributed across abdominal segments II-VI.
Second instar Similar in shape to adult female. Legs and antennae shorter. Hair-like microtrichia present on abdomen
D.J. WILLIAMS
only. Short sensory setae extending to abdominal seg- ment VII. Circulus large as in adult female.
Adult Male (Fig. 15).
Body of adult male degenerate, elongate, slender, largest specimen 1.35 mm long, 0.50 mm wide, widest at mesothorax, head and thorax rounded, then gently tapering to rounded posterior end, base of abdominal segment VIII about 150 mm wide. Area between an- tennae on venter tessellated. Antennae short, almost conical, 2-segmented, 37.5-45.0 mm long, situated near ventral head margin. Legs squat, coxae wide; trochanter + femur fused, unsegmented, stout, about 67.5 mm long; tibia + tarsus slender, 50-57 mm long, sometimes with indistinct segmentation; claw dis- torted, stout and almost triangular, about 12.5 mm long. Genital capsule retracted inside abdomen, ven- tral slit almost square. Aedeagus about 100 mm long, stout, almost trumpet-shaped at apex. Anus oval, situ- ated at posterior end of abdomen at base of sclerotized extension to abdominal segment VIII, this extension about 1.5 mm wide and long, hook-shaped in profile, equivalent to long style in male of E. taylori.
Dorsal and ventral surfaces practically naked except for some minute setae, each with setal length consider- ably smaller than width of collar, except on dorsum of thorax where setae slightly longer.
COMMENTS. ‘There is available also a single male pupa with an adult male inside. Also available are some female pupae, one of which shows the develop- ing adult female inside.
MATERIAL
HoLoryPe. Adult 9, Malaysia, Sarawak, 1°38’N 113°35’E, associated with Acropyga sp., 25.11.1963 (A. Emerson) (ANIC).
PARATYPES. 9, Malaysia, Sarawak, same data as holotype, 1 adult ?, 7 adult dic’ (ANIC), 2 adult 92, 7 adulto'o' (BMNH), 2 2nd instars (ANIC), 2 2nd instars (BMNH), 5 pupae 9? (ANIC), 3 pupae 99 (BMNH), 3 pupae dc’ (ANIC), 2 pupae d'o' (BMNH).
COMMENTS. This singular species has an unusual shape, evenly tapering posteriorly from the prothorax. It belongs to a group collected within the tropics and is extensively clothed in hair-like microtrichia and minute setae. E. sarawakensis comes closest to E. sulawesicus from which it differs mainly in possessing shorter and fewer stout sensory setae on the cephalothorax, mostly 17.5—22.5 um long. In E. sulawesicus, these setae are noticeably longer, 45-50 um long and more numerous.
Most of the differences are small but both species are known from widely isolated localities and at present this separation is justified.
The epithet is based on the name of the locality and the Latin suffix ‘-ensis’ indicating origin.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 19
Eumyrmococcus scorpioides (De Lotto) (Fig. 16)
Xenococcus scorpioides De Lotto, 1977: 33. HOLOTYPE, SouthAfrica, Cape Province, Jacobs Bay (SANC) [examined].
Eumyrmococcus scorpioides (De Lotto), Williams, 1993: 217; Ben-Dov, 1994: 152.
DESCRIPTION
Appearance in life not recorded. Adult female on microscope slide elongate, body membranous except for moderately sclerotized abdominal segment VIII, largest specimen 1.55 mm long, 0.77 mm wide, widest at mesothorax, head and thorax rounded, with a con- striction between abdominal segments III and IV, narrowing abruptly, remainder of abdomen tapering with a further constriction between abdominal seg- ments VII andVIII; segment VIII about 170m wide at base, posterior end rounded. Positions of each anal lobe with 2 stout ventral setae and | dorsal seta, each about 800 um long, forming a group of 3.Antennae slender, 4- segmented, each 175-190 um long. Legs well developed; hind trochanter + femur 170—180 um long, hind tibia + tarsus 150-162 um long, claw about 27.5 um long, stout, 12.5 um wide at base. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.87—0.92. Ratio of lengths of hind tibia to tarsus 1.41— 1.61. Outer setae on legs sensory, slightly swollen at tips; inner setae flagellate. Labium 147-150 um long, about as long as clypeolabral shield, 50-65 um wide; ratio of length to width 2.26—3.00. Circuli numbering 2, situated anteriorly on abdominal segments II and II but within borders of segments; circulus on abdominal segment II about 25 1m in diameter, on segment III about 15 {1m in diameter, both cup-shaped. Anal ring 57-60 tum wide, with 6 slender setae; posterior pair each about 87 [im long, anterior 2 pairs shorter but curled in available specimens.
Dorsal surface with crowded flagellate setae on abdomen, those at posterior end of abdominal segment Vill each about 75um long, on abdominal segment VII mostly about 37 Um long, anteriorly about 25 um long. Head and thoracic setae mostly sensory, with slightly swollen tips, 12.5—25.0 um long but some flagellate setae present mingled with the sensory setae.
Ventral surface with similar setae to those on dor- sum, crowded and flagellate as far forward as abdominal segment III but a few sensory setae also present on abdominal segment III. Anteriorly, setae mostly sensory with swollen tips except in medial area of thorax where they are mostly flagellate but a few sensory setae also present.
MATERIAL
South Africa: Cape Province, Jacobs Bay, in nests of Acropyga (Malacomyrma) arnoldi Sanschi, 7.viii. 1969 (A.J. Prins).
COMMENTS. In possessing 4-segmented antennae and in the general distribution of body setae, this species is related to E. corinthiacus. Most dorsal and ventral setae on the head and thorax of E. scorpioides, how- ever, are sensory with slightly swollen tips, whereas in E. corinthiacus they are all flagellate. Moreover, the long setae on the anal lobes of E. scorpioides are each over 800 ttm long and those of E. corinthiacus are thicker but at most only about 200-230 um long. E. scorpioides differs from all other known species of Eumyrmococcus in possessing stout claws, those in other species being much more slender.
The accompanying illustration is based on the holotype and two paratypes kindly made available for this work by Ian Millar, Plant Protection Research Institute, Pretoria, South Africa.
Eumyrmococcus smithii Silvestri
(Fig. 17)
Eumyrmococcus smithii Silvestri, 1926: 273; Williams, 1970: 138; 1978: 63; Ben-Dov. 1994: 152. LECTOTYPE, China: Macao, [on roots of plants, attended by the antAcropyga (Rhizomyrma) sautori Forel], designated by Williams, 1978: 63 (IEAUN) [examined].
COMMENTS. ‘This species was described and illus- trated in detail by Williams (1970) from material collected in various localities in China and Japan. Dr M. Terayama has kindly made available some fresh material collected in Okinawa and a new illustration is presented here based on this material.
The most important characters are the 2-segmented antennae, the absence of a circulus, long stout setae in groups of three on each apparent anal lobe, each seta about 500 tm long in the species illustrated but up to 700 Lim long in other specimens. The anal ring pos- sesses 6 setae, the 2 anterior pairs short and slender and the posterior pair stout, about as long as the anal setae. All the body setae are short, slender and flagel- late, occupying all the dorsal surface except for intersegmental areas. A striking character, not seen in any other species, is a long wide extension on each hind coxal process.
In lacking any sensory setae with blunt or swollen tips, E. smithii is related to E. kolombangarae and E. Kusiacus, described here from the Solomon Islands. It differs from both of these species in lacking a circulus.
MATERIAL China: Macao, in nest of Acropyga (Rhizomyrma) sauteri. Taiwan: Taichu, on sugarcane, 24.11.1933 (M. Yanagihara). Japan: Ryukyu Is, Okinawa Is, in nest of A. (Rhizomyrma) sauteri, 15.vii.1984(M. Terayama); without locality, ex coll. R. Takahashi.
The species was also recorded from Shanghai by
20
Silvestri (1926, 1927) and from Taiwan by Takahashi (1934).
Eumyrmococcus sulawesicus sp. 0. (Fig. 18)
DESCRIPTION
Adult female on microscope slide narrowly elongate, membranous, largest specimen 1.50 mm long, 0.60 wide, widest at mesothorax; head and thorax rounded then constricting slightly between abdominal segments VII and VIII, abdominal segment VIII 162 um wide at base. Position of each anal lobe with 2 stout dorsal setae and | ventral seta, forming a group of 3, each 510-525 um long. Antennae placed well on to dorsum, each 275-295 wm long, with 2 segments, second tapering, 215-232 um long. Legs well developed, slender; hind trochanter + femur 200-205 tm long, hind tibia + tarsus 182.5—205.0 um long, claw slender, about 27.5 um long, 7.5 tum wide at base. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.90-1.02. Ratio of lengths of hind tibia to tarsus 0.87—0.95. Leg setae short and sensory on outer edges, each with swollen tip, flagellate on inner edges. Labium about 142.5 um long, longer than clypeolabral shield, 95 Um wide, ratio of length to width 1.5. Circulus large, round to slightly elliptical. Anal ring 72-78 wm wide with 6 setae; anterior pair slender, each about 45 um long, second pair each about 75 um long, posterior pair stout, each about 475 um long, resembling anal lobe setae with which they are some- times intertwined.
Dorsal surface entirely covered with minute slender setae and fine microtrichia. Setae on abdominal seg- ment VIII mostly about 15 um long, anteriorly about 6-10 um long, many of longer setae with larger setal collars, density less towards head and thorax. Long sensory setae, each with swollen tip, 45-50 um long, present on head and thorax. Microtrichia dense, hair- like, short and stouter on abdominal segment VIII, very slender anteriorly. Setae and microtrichia absent from intersegmental areas.
Ventral surface with similar setae and microtrichia to those on dorsum, covering entire surface except for intersegmental areas. Long, stout sensory setae, as on dorsum, present on head and thorax, fairly numerous, becoming shorter, 17.5-25.0 um long, towards ante- rior abdominal segments except for one or two, each about 32 um long near circulus. Sensory setae towards abdominal segment V mostly about 15m long, some- times blunt, without swollen tips. Setae stout and flagellate on medial area of thorax.
MATERIAL HOLOTYPE. Adult 2, Indonesia, Sulawesi, Sulawesi Utara, Toraut forest, Dumoga-Bone National Park,
with Acropyga (Atopodon) sp., 15.11.1985 (R.HLL. Disney) (BMNH).
D.J. WILLIAMS
PARATYPES. Indonesia, Sulawesi, same data as holotype, 3 adult 9? (BMNH), 6 2nd instars (BMNH).
COMMENTS. This species, covered in hair-like microtrichia, is noticeably elongate with the antennae placed well on the dorsum of the cephalothorax. It is closely related to E. sarawakensis in most characters but the stout sensory setae on the head and thorax are longer, up to 50 um long, and more numerous. In E. sulawesicus these setae are at most about 22.5um long.
A few second instars are also available. They differ from the adult female in having much fewer minute setae and microtrichia, but the stout sensory setae are more numerous, extending posteriorly as far as ab- dominal segment VII.
The epithet is based on the country of origin, Sulawesi, and the Latin suffix ‘-icus’ meaning ‘per- taining to’.
Eumyrmococcus taylori sp. n. (Figs 19-22)
DESCRIPTION
Appearance in life not recorded. Adult female (Fig. 19) on microscope slide elongate, narrow, membranous except for posteriorend of body moderately sclerotized; longest specimen 1.30 mm long, 0.50 mm wide, widest at mesothorax; constricting between abdominal seg- ments IV and V, widening at abdominal segment VI, then narrowly tapering between abdominal segments VIL andVIIL, base of abdominal segment VII 117-125 um wide; posterior end of body rounded, projecting beyond actual anal lobes, these each with 2 ventral setae and | long stout seta forming a group of 3, each seta about 315 um long. Antennae each situated on dorsal margin when flattened on slide, 310-390 um long, with 2 segments, second segment long, about 290-330 um long, curved and tapering. Legs well developed, slen- der, hind trochanter + femur 170-210 Lm long, hind tibia + tarsus 150—190 Um long, claw slender, about 27 uum long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.88—0.90. Ratio of lengths of hind tibia to tarsus 1.14—1.25. Tibia swollen towards distal end then tapering. Tarsus swollen then tapering to narrow distal end, about as wide as base of claw. Leg setae all flagellate. Labium aboutas long as clypeolabral shield, 117-120 um long, 77.5—82.5 tum wide; ratio of length to width 1.41—1.51. Circuli numbering | or 2, situated within borders of abdominal segments II and III; if only 1 circulus then present on abdominal seg- ment III; circulus on abdominal segment II about 8.75 tum wide, on abdominal segment III about 11.25 um wide, each circulus deeply cup-shaped. Anal ring 53— 57 tum wide, with 6 setae; anterior pair each about 67 uum long, second pair each about 1 12 um long, posterior pair thicker, each about 275 um long, resembling the apical setae.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI PAN |
Dorsal surface with crowded flagellate setae as far forward as abdominal segment II, mostly about 17.5— 25.0 um long on abdominal segment VIII, those on anterior segments each about 15 tm long, very slen- der. Minute swollen setae present, each about 3.75 um long, sparse, mingled with the flagellate setae on abdominal segments I-IV. Setae anterior to abdomi- nal segment II mostly thick and flagellate, 25-50 um long, not numerous.A few minute swollen setae present on head, each about 3.5 um long, slightly narrower than those on abdomen.
Ventral surface with similar setae to those on dor- sum, crowded as far forward as abdominal segment II, with a few minute swollen setae on abdominal seg- ments II-VI. Anteriorly a few thicker flagellate setae present as on dorsum, and a few minute swollen setae also present on head and around anterior spiracles.
Second instar (Fig. 20)
Body pyriform, 1.05—1.10 mm long, 0.45—0.58 mm wide, widest at about mesothorax, tapering abruptly to abdominal segment IV then gradually to posterior end of body; base of abdominal segment VIII 75—92 Um wide; posterior end of body projecting beyond actual anal lobes, each lobe with | dorsal and 2 stout ventral setae each 375-470 ttm long. Antennae each 290— 350 um long, with 2 segments; first 40-50 Um long, second curved, with setae of various lengths, longest at distal end about 100 Lm long. Legs well developed; hind trochanter + femur 155—195 um long, hind tibia + tarsus 157.5—185.0 um long, claw slender, about 22.5 uum long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 0.94—1.04. Ratio of lengths of hind tibia to tarsus 1.03—1.25. First pair of legs longer, trochanter + femur 170-200 tm long, tibia + tarsus 185—207 um long. Labium broad, 105—115 tm long, longer than clypeolabral shield. Circulus present within borders of abdominal segment III, 7.50-11.25 Um in diameter, deeply cupped. Anal ring 35-40 um wide, with 6 setae; 2 anterior pairs slender, each about 75 um long, posterior pair stout but not so stout as anal lobe setae, each about 350 um long.
Dorsal surface with long stout setae 95-120 um long at posterior edges of abdominal segments I[I— VII: other abdominal setae, each about 25 Lm long, present on abdominal segment VIII, and others, each about 20 tum long, at anterior edges of abdominal segments IV—VII; slenderer setae on these segments mostly about 15 um long. Other setae on head, thorax and abdominal segments I-III, each 20-55 tum long, not numerous. Minute clavate setae present on ab- dominal segments II-IV, and others, about same size but with smaller collars, present on head and lateral area of metathorax.
Ventral surface with similar setae to those on dor- sum. Abdominal segments V and VI each with a few elongate sensory setae; abdominal segments III and IV
each with a few shorter sensory setae. Minute clavate setae present anterior to each spiracle and posterior to each second spiracle.
COMMENTS. Material from Baroalba Spring pos- sesses legs and antennae slightly longer than that from Sawcut Gorge but the proportions of the segments are about the same. At present, specimens from both areas are treated here as conspecific.
Female pupa (Figs 1B, 21A)
Body membranous, elongate-pyriform, 0.90—0.99 mm long, 0.38-0.43 mm wide, widest at mesothorax, ab- domen gently tapering, posterior end projecting only slightly, segmentation distinct on abdomen. Antennae situated on dorsal margin, 2-segmented, curved, 262— 270 um long, apex rounded. Legs tapering to pointed developing claws, 325-390 um long, with faint seg- mentation. Labium distinct, 100—120 um long, longer than clypeolabral shield. Anal ring dorsal, at apex of abdomen. Spiracles present.
COMMENTS. Specimens are available showing the adult female inside, almost ready to emerge (Fig. 1B).
Adult Male (Fig. 22)
Body of adult male elongate-pyriform, 104 Um long, 40 um wide, widest at mesothorax, head and thorax rounded, tapering abruptly to anterior end of abdomi- nal segment IV, remainder of abdomen narrow, tapering gradually, base of abdominal segment VII 90 um wide. Apparent anal lobes each with | stout dorsal seta and 2 ventral setae, each 200-250 ttm long, forming a group of 3. Antennae 2-segmented, 310—400 tm long; second segment curved, 270-350 1m long, longest setae about 90 um long. Legs well developed, slender; hind trochanter + femur 155—175 um long; hind tibia + tarsus 147.5— 175.0 um long; claw slender, 25 um long, 5 um wide at base. Ratio of lengths of hind tibia + femur to trochanter + femur 0.94-1.01. Ratio of lengths of hind tibia to tarsus 1.03—1.33. First legs longer; trochanter + femur 180-210 Lim long, tibia + tarsus 180-200 um long. Mouthparts represented by tentorium, remains of clypeolabral shield with a few setae, and a small vestigial labium. Genital capsule almost triangular, 65 um wide, length about the same, venter of penial sheath rounded apically, ventral slit rounded near base then widening slightly apically; basal ridge of penial sheath well defined ventrally; aedeagus long and pointed, 250—262 1m long. Dorsal aspect with an apparent anal ring and anal ring setae at posterior edge of abdominal segment VIII; 2 anterior pairs of setae each about 45 um long, posterior pair thicker, each about 58 Lim long, situated near curved ends of anal ring and lateral to large, almost triangular anus, about 30 um wide. Dorsal part of capsule ex- tending from anus as a slender elongate style, about 125 um long, shorter than aedeagus, slightly expanded
22
apically then pointed, tip bearing a few spicule-like extensions; base of style with 2 pairs of short setae.
Dorsal surface of body with short, flagellate setae, each about 12.5—15.0 um long, on abdominal segment VIII. Anteriorly on all segments including head, with transverse rows of long, slender flagellate setae 62— 70 um long, these sometimes absent; posterior abdominal segments also with a few short flagellate setae. Minute sensory setae, each with slightly swollen tip and scarcely longer than a setal collar, present in moderate numbers across segments anterior to ab- dominal segment VIII.
Ventral surface with similar distribution of setae to those on dorsum, the long flagellate setae sometimes absent. Minute sensory setae present as on dorsum but fewer on thorax.
Male prepupa (Fig. 21B)
Body elongate-pyriform, about 0.90 mm long, 0.40 mm wide at mesothorax; abdomen with distinct segmenta- tion, tapering to rounded apex. Antennae situated on dorsal margin, 2-segmented, tapering, apex rounded, 220-240 um long. Legs 305—330 um long, segmented, developing claws pointed. Anal ring present at apex of abdomen. Mouthparts represented by remains of clypeolabral shield and small lobes of labium. Spiracles distinct.
COMMENTS. Specimens are available showing the developing pupa inside.
Male pupa (Fig. 21C)
Body narrowly pyriform, 1.00—1.27 mm long, 0.40- 0.43 mm wide, widest at mesothorax, narrowing to abdominal segment V then tapering to developing genital capsule, posterior end elongate, pointed; seg- ments distinct on abdomen. Antennae tapering to pointed distal end, 285-300 tm long, situated on dorsal head margin. Legs tapering to pointed claws, 320-360 um long. Mouthparts represented by indis- tinct clypeolabral shield. Spiracles present. Anal ring situated dorsally towards anterior end of abdominal segment VIII.
COMMENTS. Some specimens available show the adult male complete, almost ready to emerge (Fig. 1A).
MATERIAL
HOLOTYPE. Adult 9, Australia: Northern Territory, Sawcut Gorge, 12°55’S, 132°56'E, associated with Acropyga sp., 19.xi.1972 (R.W. Taylor) (ANIC).
PARATYPES. Australia: same data as holotype, 1 adult 2, 2 adult bo' (ANIC), 1 2nd instar, 2 éo' pupae, 1 2 pupa (ANIC), | adult 9, 3 adult do’, 1 3 pupa (BMNH): same data but 13.vi.1973, 1 adult &, 4 2nd instars, 2 prepupae do’, | pupa o’, 2 pupae 9? (ANIC): 2 2nd instars, 1 prepupac' (BMNH). Baroalba Spring,
D.J. WILLIAMS
12°47’S, 132°51’E, with Acropyga sp., 17.xi.1972 (R.W. Taylor), 1 adult o', 7 2nd instars, 7 pupae 9 (ANIC), | 2nd instar, 3 pupae 99 (BMNH); same data but 17.x1.1971, 1 adult @ (ANIC).
COMMENTS. In possessing a general distribution of short crowded setae in the adult female, only as far forward as abdominal segment II, this species closely resembles E. recalvus sp. n. In E. taylori, however, all the dorsal setae on the head, thorax and first abdominal segment are flagellate except for a few minute swollen sensory setae. In E. recalvus, all the dorsal setae at the anterior end of the body are sensory and bluntly tipped or slightly swollen. Furthermore, the outer setae on the legs of E. taylori are flagellate, whereas those of E. recalvus are sensory.
There are available some second instars from all localities. These differ from the adult female in lack- ing the wide bands of dense setae across the segments and instead there are long slender setae, similar to those of the adult male.
Adult males appear to be of two forms, with or without long flagellate body setae. Most of the adult males from Sawcut Gorge possess the long setae ex- cept in one specimen. One of the adult males from Baroalba Spring also lacks these long setae as does the specimen still enclosed within the pupal instar. All specimens are here regarded as representing the same species. Such wide differences occur also in adult females of Molluscococcus fibrillae Hall, redescribed by Miller and Williams (1995). This species normally possesses extremely long setae, 812-928 um long over most of the dorsum, but there are also specimens with short setae only, 44-61 um long.
The only adult females available for study from Baroalba Spring are still within the pupal instar and they appear to be identical with adult females from Sawcut Gorge. If there is evidence in future that the two forms of adult male discussed here, represent two distinct species, then the matter can easily be cor- rected.
The species is named after the collector, R.W. Taylor, CSIRO, Canberra, Australia, who sent many of the mealybugs discussed in this work.
Xenococcus Silvestri
Xenococcus Silvestri, 1924: 312; Williams, 1978: 63.
TYPE SPECIES. Xenococcus annandalei Silvestri, by original designation and monotypy.
DESCRIPTION
Body of adult female broadly oval, membranous, ab- domen tapering abruptly to narrow sclerotized apical segment. Anal lobes not developed, recognizable by inner ventral grooves, position of each lobe with a group of 3 long, stout setae, 2 ventral and | dorsal.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 23
Anal ring protruding between anal lobes, represented by crescentic dorsal band without cells, with 8 anal ring setae, the 2 anterior pairs of setae slender, the third pair thicker and longer, detached from ring, the fourth pair ventral in position, about as long as anal lobe setae. Antennae 4-segmented, placed on dorsal mar- gin, tapering, about as long as body with strong articulation between first and second segments. Legs well developed, long and slender; claw elongate, slen- der. Labium elongate, longer than wide, setae on upper surface well spaced. Circuli present, round and shallowly cupped. Dorsal body setae minute and abun- dant, extending to lateral ventral margins on thorax. Ventral setae mostly long and stouter. Sickle-shaped setae usually present on thorax. Eyes absent. Ostioles absent. Pores and ducts absent.
COMMENTS. In life, the abdomen of Xenococcus curls slightly to the dorsum as in Eumyrmococcus. The genus differs from Eumyrmococcus in possessing very long antennae with well developed articulation be- tween the first and second segments. Although the abdomen tapers, it narrows abruptly to a narrow ab- dominal segment VIII and the body is widest at about the metathorax. In Eumyrmococcus, the whole cephalothorax is dilated and the abdomen tapers gradu- ally. Furthermore, only the dorsal abdominal setae in Xenococcus are short and crowded, those on the venter of the abdomen, although numerous, are long and stouter, similar to the medial ventral setae on the head and thorax. The ventral abdominal setae in Eumyrmococcus are always short and crowded, simi- lar to the dorsal abdominal setae.
At present, two species are recognized, always as- sociated with the ant genus Acropyga. The female possesses a pupal instar and the adult male of one of the species is described on p. 24.
Key to Species of Xenococcus (Adult Females)
1 Legs with tibia shorter than tarsus. Antennae each about 860-1090 Lm long ou... ee eeeeeeeeeeeee acropygae (p. 23)
— Legs with tibia about twice as long as tarsus. Antennae about 1480-1550 um long............... annandalei (p. 25)
Xenococcus acropygae sp. n.
(Figs 23-26)
Xenococcus annandalei Silvestri, Williams, 1985: 390; Williams and Watson, 1988: 221 (mis-identifica- tions).
DESCRIPTION
Adult female (Fig. 23) on microscope slide pyriform, narrowing at abdominal segment VII; 1.27—1.65 mm long, 0.74—1.08 mm wide, widest at about metathorax;
body membranous except for sclerotized abdominal segment VIII, posterior end of body projecting beyond anal lobes; base of abdominal segment VIII 180- 220 um wide; anal lobes each with | dorsal and 2 ventral setae, all stout, forming a group of 3, each 875— 920 um long. Antennae conspicuous, each 860-1090 um long, with 4 segments; segment | 170- 230 um long (type series 90-120), segment 2 280-300 tum long (type series 330-360), segment 3 90-130 um long (type series 90-120), segment 4 200- 300 um long (type series 240-300); segment | 170-230 lim wide, segments progressively narrowerer to seg- ment 4 60-100 tum wide, segment 4 tapering. All antennal segments with long stout setae, segment 2 with about 30 setae on ventral surface; longest setae on segment 4 400 Um long, setal bases 3—5 um wide with correspondingly wide setal collars. Segments | and 2 strongly articulated with grooves at distal end of seg- ment | and minute projections at proximal end of segment 2. Tip of segment 4 with a pair of peg-like setae each 10.0—12.5 um long. Legs well developed; hind trochanter + femur 270-380 um long (type series 290-340), hind tibia + tarsus 280-415 um long (type series 310-360), claw slender about 35—45 um long. Ratio of lengths of hind tibia + tarsus to hind tro- chanter + femur 1.01—1.13. Ratio of lengths of hind tibia to tarsus 0.66—0.95; tarsus always longer than tibia. Hind tarsus swollen then narrowing distally. All legs with long stout setae, the longest at distal end of femur, 250-300 um long. Labium 165-220 um long, longer than clypeolabral shield, setae well separated. Circuli normally numbering 2 within borders of ab- dominal segments If and II, each 27.5—47.0 ttm in diameter, with rim projecting from surrounding derm, inner part cup-shaped, shallow. Occasionally a small third circulus present on abdominal segment IV. Anal ring 77.5—85.0 um wide with 8 setae; 2 anterior pairs slender, each about 130 um long, 2 posterior pairs stout, each about as long as anal lobe setae, | pair on dorsum and the posteriormost pair usually placed on venter.
Dorsal surface densely covered with slender setae except in intersegmental areas and on head, mostly curved, each about 15 tum long towards anterior edges of segments, and about 20 um long towards posterior edges of segments; a few longer setae, each about 25 um long with larger collars, present across all segments; abdominal segment VIII with fewer but longer setae. Sickle-shaped setae usually present on thorax and anterior abdominal segments but some- times absent entirely, each with slender tip. Setae near each antennal base slender, stouter on head margin.
Ventral surface with stout setae on abdomen and medial areas of head and thorax, 40-165 «1m long except for some on margins of abdominal segment VIII, each about 300 um long. Short slender setae as on dorsum, present around margins of anterior ab-
24
dominal segments and thorax. A few sickle-shaped setae usually present near margin of thorax.
COMMENTS. This species differs from X, annan- dalei mainly in the lengths and ratios of the tibiae and tarsi. In X. acropygae, the tarsus is longer than the tibia but in X. annandalei, the tibia is about twice as long as the tarsus. Furthermore, the setal bases on the antennae of A. acropygae are mostly narrower than those of X. annandalei. Moreover, the antennae and legs of A. acropygae are shorter than those of A. annandalei.
First instar (sex not determined) (Fig. 24)
COMMENTS. ‘This instar was described by Williams (1978) from specimens collected in India, Mysore, on coconut roots under the name X. annandalei. Speci- mens from southern Asia and Australasia agree with these first instars but show wider variation. The body varies from 0.70—1.05 mm long and 0.27-—0.58 mm wide. Antennae 4-segmented, 740-880 um long, with long stout setae, the longest on segment 4 about 250 um long. Hind trochanter + femur 175—250 um long, hind tibia + tarsus 215-270 um long. Ratio of lengths of hind tibia + tarsus to trochanter + femur 1.08-1.27. Ratio of lengths of hind tibia to tarsus 0.58—0.74, tarsus always noticeably longer than tibia. All segments with long stout setae, longest at distal end of femur about 160 um long. Anal ring with 4 pairs of setae; 2 anterior pairs slender, each about 60— 110 um long, third pair stouter, 295-360 Um long, fourth pair stoutest, 535-750 um long, resembling the anal lobe setae. Anal lobe setae, stout, usually curled when prepared on slides, 675—700 Lm long, forming a group of 3 on each anal lobe, | on venter and 2 on dorsum. Body setae short and curved on head and thorax, each 12—15 tum long, accompanied by a few sickle-shaped setae on dorsum. Other body setae 40-88 um long except on margins where they are 100-120 um long.
A new illustration has been prepared for this work based on specimens collected at the type locality, Sulawesi.
Third-instar female (pupa) (Fig. 25A)
Body broadly oval, 0.84—0.90 mm long, 0.55—0.67 mm wide, membranous, with only faint signs of segmenta- tion at posterior end of abdomen. Antennae dorsal, 660-760 tm long, tapering. Legs tapering, 425— 530 um long, segmentation faint. Labium distinct, 135-145 um long, longer than clypeolabral shield. Spiracles distinct.
COMMENTS. The female pupal instar is always rec- ognizable by the well developed labium and the long dorsal antennae. Some specimens available have the developing adult female still within the pupal instar.
D.J. WILLIAMS
Adult male (Fig. 26)
Body normally curved ventrally and distorted when prepared on microscope slides. When flattened, broadly oval, about 1.0 mm long, 0.70 mm wide, sides sub-parallel, widest at about abdominal seg- ments II and V. Head and abdominal segment I sclerotized dorsally except at anterior end surround- ing antennae; ventrally, sclerotized laterally on thorax. Antennae situated on dorsal membranous area, each 1-segmented, tubercle-like, with 6 stout setae, each 30-35 um long. Legs placed well anterior on body, of an unusual shape, squat and robust. Coxae large. Tro- chanter + femur stout, about 70 Lim long, fused except for faint line apparent in some specimens, trochanter with the usual 2 pairs of sensory pores. Tibia + tarsus much narrower, tapering, about 75 um long. Claw about 20 Lim long, unusual, with well developed wide base, actual claw pointed and slender with apparently a pair of stout blunt digitules. Genital capsule about 290 um long, 110 um wide, mostly internal, the external venter of penial sheath about 170 um long; ventral slit indistinct, at distal end of sheath; tip of sheath with a few minute setae. Aedeagus strongly sclerotized, elongate. Base of genital capsule dorsally with a pair of lobes, each with a group of 3 stout setae, 30-37 ttm long, and usually 4 short setae, the lobes situated on each side of a minute projecting plate; anus, when viewed laterally, opening under the plate; entire area normally lying between the genital capsule and dorsal surface of abdomen when flat- tened on slide.
Body setae of 2 types.A band of setae present across anterior edges of venter of abdominal segments II and Ill, each seta minute and shorter than diameter of heavily sclerotized collar. A minute type of seta with membranous collar, present across dorsal and ventral segments; setae in a row near ventral edge of abdomi- nal segment IV either with membranous or sclerotized collars.
Third-instar male (prepupa) (Fig. 25B)
Body almost rotund, 0.78—0.86 mm long, 0.66— 0.70 mm wide. Segmentation discernible on dorsum and venter of abdomen. Antennae dorsal, tapering to blunt apex, 185-260 um long. Legs 180-240 um long, tapering, with faint segmentation. Remains of clypeolabral shield and labium present. Spiracles well developed.
COMMENTS. The prepupal instar is recognizable by the dorsal developing antennae, only about one-third the length of those of the female pupa. There are faint signs of the labium and clypeolabral shield.
Fourth-instar male (pupa) (Fig. 25C) Body broadly oval, 0.86—0.98 mm long, 0.65—0.75 mm wide, with signs of segmentation on abdomen. Anten-
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI Pa)
nae reduced to small triangular flaps on venter of head, 25-30 uum long, 50-65 tum wide. Legs 155-200 um long, stout and tapering to sclerotized points or devel- oping claws, segmentation not discernible. Small area of clypeolabral shield present only. Spiracles distinct.
COMMENTS. In the male pupa the legs are usually shorter than those of the prepupa but the lengths sometimes overlap. The most distinctive pupal charac- ter is the shape of the antennae which are short, triangular and present on the anterior edge of the head to contain the short developing antennae which in the adult male lie dorsally.
MATERIAL
HOLOTYPE. Adult 2, Indonesia, Sulawesi, Sulawesi Utara, Dumoga-Bone National Park, Toraut forest, with Acropyga (Acropyga) acutiventris Roger, 19.111. 1985 (R.H. Disney) (BMNH).
PARATYPES. Indonesia, Sulawesi, same data as holotype but (R.H. Disney or R.H. Disney & J.H. Martin), 3 adult 99 (BMNH), 10 adult éo&' (BMNH), | adults’ (ANIC), | adulto' (MNHN), | adulto’(USNM), 18 Ist instars, 24 pupae 92, 4 prepupae do’, 3 pupae do (BMNH).
Non-type material (all 99). Indonesia, Krakatau, Anak Krakatau, Rakata (in mandibles of A. acutiventris). India, Mysore, on roots of Cocos nucifera, 21.vi.1937 (T.V. Subramanian) (first instar); Karnataka, Bangalore, on roots of Vitis vinifera, 18.1.1996 (A. Virakramath & B.K. Rajagopal). Singa- pore; Botanical Gardens, in soil under wood, with A. acutiventris, 21.v.1968. Philippine Islands, Palawan (spirit material). Malaysia, Sabah, Tawau, Quoin Hill, with A. acutiventris, 10.vi.1968. Papua New Guinea, Morobe Province, 16 km N.W. Lae, in soil under wood, with A. acutiventris, 12.vi.1972 (R.W. Taylor); Wau, McAdam Park, in rotting log, with A. acutiventris, 14.vi.1972 (R.W. Taylor); Bulolo (spirit material, with A. acutiventris and A. (Atopodon) ambigua): East Sepik Province, Angoram, from rot- ting log, 2.vi.1972 (R.W. Taylor); Yawasora, nr Wewak, from rotting log, 4—6.vii.1972 (R.W. Taylor); Hayfield, nr Maprik, in soil under wood (all with A. acutiventris): West Sepik Province, Pes, nr Aitape, in soil under wood with A. acutiventris, 8—9.vi1.1972: Northern Province, Kokoda, associated with dying Theobroma cacao, ix.1990 (G. Lockwood); nr Kokoda, in soil under wood, with A. acutiventris, l.vi.1972 (R.W. Taylor). Solomon Islands, San Cristobal, Guadalcanal, Nggela (spirit material), with A. acutiventris and A. (Rhizomyrma) lauta Mann. Australia, Queensland, Iron Range, 12°42’S 143°18’E, 9-15.vi.1971 (R.W. Taylor & J. Feehan); Waugh, from rotten log, 11.vi.1962 (R.W. Taylor); Cape York Peninsular, Bemaga, from mandibles, xii.1983 (J. Sedlack); Finch Hatton Gorge, 21°05’S
148°38’E, 11.xi.1976 (R.W. Taylor & TA. Weir); Josephine Falls (spirit material): Northern Territory, Baroalba Spring, 12°47’S 132°51’E, in soil under wood, 16, 17, 20.x1.1972, 13.vi.1973 (all with A. acutiventris).
COMMENTS. Specimens recorded under the name X. annandalei from Hong Kong, Vietnam and Penang by Williams (1978) are probably this species and await verification.
Xenococcus annandalei Silvestri
(Fig. 27)
Xenococcus annandalei Silvestri, 1924: 312; 1926: 275; 1927: 253. LECTOTYPE 2, INDIA: Barkuda I. [Chilka (Chilika) Lake, Madras District (now Orissa Province), on roots of Ficus obtusa, with antsAcropyga acutiventris Roger] (IEAUN, Portic1) designated by Williams 1978: 66 [examined].
COMMENTS. Williams (1978) described the adult female in detail and the illustration is reproduced here with slight modification. Some further notes are now added to distinguish the species from X. acropygae.
The antennae are 1480-1550 [tm long, longer than in any specimen of X. acropygae so far studied. In X. acropygae, the range is 860-1090 um long. Further- more, although the distribution of the antennal setae are about the same in the two species, there are many more shorter and more slender setae in X. annandalei. The bases of the setae in the antennae of X. annandalei vary in width from 2.5 um to 7.5 um with a corre- sponding difference in size of the setal collars, whereas in X. acropygae there is less variation, with the width of the setal bases 3—5 tm. The legs of X. annandalei differ from those of X. acropygae. In X. annandalei, they are longer, with the hind trochanter + femur 440-480 um long and the hind tibia + tarsus 530-560 uum long. The ratio of lengths of hind tibia + tarsus to those of the hind trochanter + femur is 1.16—1.18 and the most striking difference is the ratio of the hind tibia to tarsus, with the tibia always about twice as long as the tarsus. Besides, the tibiae and tarsi are fairly uniform in width for most of their lengths, although the tarsi narrow and taper distally. In X. acropygae, the tarsus is always longer than the tibia and the tibia is swollen before narrowing abruptly to a long slender distal end.
The short flat setae on the dorsum discussed by Williams (1978) are, in fact, sickle-shaped in profile, similar to those of X. acropygae.
The third instar discussed by Williams (1978) is now known to be the second instar (Williams, 1988). It is also evident that the first instar discussed by Williams (1978) from Mysore, on coconut roots, is the first instar of X. acropygae.
26
MATERIAL
At present the species is only known from India, Orissa Province, Barkuda I., on roots of Ficus obtusa and F: religiosa.
REVIEW OF GENERA IN THE RHIZOECINAE
In addition to Eumyrmococcus and Xenococcus, al- ready discussed, the following genera have been included in the subfamily at one time or another and their current status is discussed.
Tang (1992) erected the subtribes Rhizoecina, Prorhizoecina and Pseudorhizoecina for some of the genera, but this action seems unnecessary. These names, nevertheless, remain available.
The opportunity is taken to describe a new genus from southern Asia because it resembles Eumyrmococcus superficially but is clearly more re- lated to Rhizoecus.
Brevicoccus Hambleton
Brevicoccus Hambleton, 1946a: 10; Williams & Granara de Willink, 1992: 68. Type species: Brevicoccus clavisetosus Hambleton, by original designation and monotypy.
DIAGNOsIS. Body broadly oval, anal lobes poorly developed, each without any development of long setae. Antennae placed close together, strongly genicu- late, 4-segmented, last segment triangular. Legs well developed; claws long and narrow, Anal ring fairly simple, with a few cells and 8—16 short knobbed setae. Ostioles present. Eyes absent. Cephalic plate present. Body setae short, clavate. Trilocular pores present. Multilocular disc pores present on venter. Minute oval disc pores present.
COMMENTS. The characters of the anal ring and antennae are distinct and the genus is recognized by most workers as belonging to the Rhizoecinae. This is a monotypic genus, known only from Brazil, and the single species is reported to live on roots of Gramineae and Cyperaceae in ant tunnels.
Capitisetella Hambleton
Capitisetella Hambleton, 1977: 40; Williams & Granara de Willink, 1992: 71. Type species: Pseudorhizoecus migrans Green, by original des- ignation and monotypy.
DIAGNOSIS. Body almost pyriform, abdomen con-
D.J. WILLIAMS
stricted slightly near posterior end. Anal lobes not developed, without differentiated anal lobe setae. An- tennae placed fairly wide apart, 3-segmented, the third segment tapering, antennal setae clavate. Legs stout, with capitate setae, claws elongate. Anal ring heavily sclerotized, with a few elongate cells and 6 setae. Ostioles absent. Cephalic plate absent. Eyes absent. Body setae all capitate. Trilocular pores present. Mul- tilocular pores absent. Large granular discoidal pores present at posterior end of body.
COMMENTS. The combination of a constriction at the posterior end of the body, reduction of antennal segments to three, the capitate body setae and the absence of ostioles, link this South American genus with Eumyrmococcus. In Capitisetella, there are a few elongate cells on the anal ring and these are absent in Eumyrmococcus. Furthermore, Capitisetella lacks the abundant setae on the abdomen, present in Eumyrmococcus or even in Neochavesia, also possi- bly related to Capitisetella. The single species lives in association with Acropyga (Rhizomyrma) paramari- bensis (Biinzli, 1935). At present, the genus is accepted in the Rhizoecinae.
Geococcus Green
Geococcus Green, 1902: 262; Williams, 1969b: 508. Type species: Geococcus radicum Green, by origi- nal designation.
DIAGNOSIS. Body elongate to oval, anal lobes well developed, sclerotized, each terminating in a stout, spine-like seta. Antennae geniculate, 6-segmented. Legs well developed, inner edges of tibia and tarsus with stout setae, claws long and slender. Anal ring with large cells and 6 setae. Ostioles present. Circuli present, raised. Cephalic plate present. Multilocular disc pores present. Trilocular pores present. Large trilocular pores present, usually much larger than the normal trilocular pores. Body setae fairly abundant, short and flagellate.
COMMENTS. The genus is widely accepted in the Rhizoecinae. At present six species are assigned to the genus from South-East Asia, southern Asia and the Indo-Australian Region but there are many species awaiting to be described. The genus is well known for the spread of G. coffeae throughout much of the trop- ics.Although this species was described from Surinam, where it is associated with Acropyga (Rhizomyrma) paramaribensis on coffee roots and other plants, it is undoubtedly native to southern Asia.
The peculiar large trilocular pores, always present, may be modified tritubular cerores. It is often difficult to study the anal ring because it is concealed between the sclerotized anal lobes.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 27
Leptorhizoecus gen. nov.
TYPE SPECIES. Leptorhizoecus deharvengi sp. n.
DESCRIPTION
Body membranous, elongate, with dilated cephalothorax, abdomen narrow, subparallel, narrow- ing abruptly to abdominal segment VIII; posterior end of body sclerotized, rounded. Anal lobes not devel- oped; setae on positions of anal lobes not differentiated from other setae on abdominal segment VIII. Anal ring ventral, crescentic, with a single row of minute elon- gate cells and 6 spine-like setae. Antennae placed close together on venter of head margin, 6-segmented. Legs well developed; with thick pointed setae on inner edge of tibia and tarsus. Claw elongate and slender. Labium narrow, longer than wide. Ostioles and circu- lus present. Body setae flagellate, mostly short and fairly numerous. Trilocular pores present on dorsum and venter. Quadrilocular pores present next to circu- lus. Tubular ducts absent.
COMMENTS. This genus is typically rhizoecine but the body shape resembles species of Eumyrmococcus in exhibiting a dilated cephalothorax. It differs from Eumyrmococcus in possessing 6-segmented antennae, placed close together, and in having ostioles and trilocu- lar pores. The ventral anal ring is unlike any so far described in the Rhizoecinae.
Leptorhizoecus is probably intermediate between the Rhizoecus and Eumyrmococcus groups and the shape suggests it is possibly attended by ants.
The name Leptorhizoecus is based on the Greek word ‘Leptos’, meaning fine and delicate, combined with the extant genus name Rhizoecus.
Leptorhizoecus deharvengi sp. n. Fig. 28.
DESCRIPTION
Appearance in life not recorded. Adult female on microscope slide elongate, cephalothorax dilated, ab- domen subparallel, widening slightly to abdominal segment VII, narrowing abruptly to rounded posterior end, body membranous except for rounded posterior half of apical segment of abdomen; 0.80-0.92 mm long, 0.27—0.36 mm wide; anal lobes not developed, anal lobe setae not differentiated from others on apical segment. Antennae placed close together on ventral head margin, each 115—135 um long, with 6 segments; falcate setae well developed on segments 5 and 6. Legs well developed; hind trochanter + femur about 92.5 tum long, hind tibia + tarsus 97.5—100.0 um long, claw slender, elongate and slightly curved, about 27.5 um long. Ratio of lengths of hind tibia + tarsus to hind trochanter + femur 1.05—1.08. Ratio of lengths of hind tibia to tarsus 0.97—1.00. Leg setae flagellate, placed well apart, inner edges of tibia and tarsus with thicker
setae. Labium 77.5-82.0 um long, about 37.5 um wide, ratio of length to width 2.06—2.18; longer than clypeolabral shield. Clypeolabral shield with 2 pairs of setae. Circulus slightly oval, about 17.5 um wide, situated within borders of abdominal segment IJ; struc- ture difficult to determine in available material. Ostioles present, posterior pair fairly well developed but with- out associated setae; anterior pair represented by indistinct slits. Anal ring ventral in position, oval to triangular, 25—30 Lim wide, with a single row of minute elongate cells in posterior arc, accompanied by 6 stout, elongate-conical setae, each 17.5 um long (cells and setae actually anterior if ring transferred to dorsum). Eyes present.
Dorsal surface with long setae, each about 60 um long, on sclerotized apical segment of abdomen; shorter setae, each about 25 um long, situated at anterior end of apical segment. Anteriorly on abdomen, setae all about 25 um long, in moderate numbers in well de- fined bands across middle of segments leaving bare intersegmental areas. Shorter setae, each 12.5—18.0 um long, present on head and thorax. All setae flagellate. Trilocular pores present, not numerous, evenly distrib- uted among the setae.
Ventral surface with similar distribution of setae to those on dorsum. Multilocular disc pores represented by a few quadrilocular pores, near circulus, each about 5 um in diameter. Trilocular pores scattered, not so numerous as on dorsum.
MATERIAL
HOLOTYPE. Adult 2, Indonesia, Sumatra, Rantau Pandan, in soil of Hevea sp., 9.vi.1991 (L. Deharveng & A. Bedos) (MNHN).
PARATYPE. Indonesia, Sumatra, same data as holotype. | adult ? (BMNH).
COMMENTS. The species is named after Louis Deharveng, Université Paul Sabatier, Toulouse, one of the collectors. Dr Deharveng has sent interesting sam- ples from southern Asia collected from Berlese funnel apparatus.
Neochavesia Williams & Granara de Willink
Chavesia Balachowsky, 1957: 158; Beardsley, 1970: . [Preoccupied by Chavesia Dollfus, 1889, Isopoda].
Neochavesia Williams & Granara de Willink, 1992: 232 [replacement name]. Type species Chavesia caldasiae Balachowsky, by original designation.
DIAGNOSIS. Body in life with tip of abdomen curled to dorsum, scorpion-like. Slide-mounted specimens with cephalothorax dilated, abdomen narrowing to a pair of protuberant, rounded anal lobes bearing many long flagellate setae. Anal ring situated at base of anal lobes, simple, with 6 or a few setae around anterior
28
sclerotized half, without cells. Antennae each with 4 or 5 segments. Legs well developed, tarsi tapering, claws long and slender. Ostioles absent. Circuli conical, each with centre cupped. Labium with 3 pairs of subapical setae. Eyes absent. Body setae short and abundant on cephalothorax, longer and in bands across abdominal segments. Trilocular pores present in type species only, each with a minute internal filament arising from centre of pore. Multilocular disc pores and tubular ducts absent.
COMMENTS. In the strong development of the cephalothorax, the simple anal ring without cells, the absence of ostioles, and the presence of cupped circuli, this South American genus comes closest to Eumyrmococcus and Xenococcus. Many species origi- nally and provisionally identified as Eumyrmococcus in South America, associated with ants of the genus Acropyga, refer to Neochavesia. The genus is here accepted as belonging to the Rhizoecinae.
Prorhizoecus Miller & McKenzie
Prorhizoecus Miller & McKenzie, 1971: 583. Type species Prorhizoecus atopoporus Miller & McKenzie, by original designation and monotypy.
DIAGNOSIS. Body oval, anal lobes not developed; positions of each anal lobe with a single normal apical seta. Antennae each with 4 or 5 segments. Legs well developed, slender, hind coxae with translucent pores. Anal ring heavily sclerotized with thick rim, cells, and 6 setae. Cerarii present on anal lobes, each cerarius with 2 conical setae and a group of trilocular pores. Ostioles present. Eyes present. Body setae short and abundant. Trilocular pores present. Multilocular disc pores present, each with 12 loculi. Tubular ducts present, each with narrow oral rim and heavily sclerotized bulbous duct.
COMMENTS. Although the antennae resemble those of Rhizoecus, the other characters, cerarii, translucent pores on the hind coxae, an anal ring with a broad rim, and peculier oral rim ducts, exclude this genus from the Rhizoecinae. The single species is known only from Mexico and feeds on grass roots. The genus may be related to Cryptoripersia Cockerell or Syrmococcus Ferris.
Pseudorhizoecus Green
Pseudorhizoecus Green, 1933: 55; Hambleton, 1977: 38; Williams & Granara de Willink, 1992: 463. Type species Pseudorhizoecus proximus Green, by original designation.
DIAGNOSIS. Body rotund; anal lobes not developed, without differentiated anal lobe setae. Antennae placed close together, each with 5 segments, tapering. Legs well developed, tarsi tapering, claws long and slender.
D.J. WILLIAMS
Anal ring terminal, wider than long, irregularly out- lined, with 2—5 short setae, a few small cells, and numerous protuberances. Ostioles, eyes and cephalic plate absent. Body setae abundant, short and flagel- late. Trilocular pores present.
COMMENTS. This unusual genus is probably related to Capitisetella but the anal ring possesses numerous protuberances and all the body setae are flagellate. The only included species has a fairly wide distribution in Central and South America and is associated with Acropyga (Rhizomyrma) rutgersi and A. paramarib- ensis.
The adult male is morphologically degenerate, with- out wings, resembling the adult female in body shape and possessing similar 5-segmented antennae. Beardsley (1970) remarked that the genitalia appeared to be similar to the genitalia of Capitistella and Neochavesia.
Pygmaeococcus McKenzie
Pygmaeococcus McKenzie, 1960: 741. Type species Pygmaeococcus morrisoni McKenzie, by original designation and monotypy.
DIAGNOSIS. Body of adult female minute, elongate, anal lobes not developed, position of each lobe with | long ventral and 2 long dorsal setae, forming a group of 3. Antennae each with 5 segments. Legs well devel- oped, with stout, spinose setae present on inner edges of tibiae and tarsi; claw long and slender, digitules clavate, equalling or surpassing claw in length. Ce- phalic plate present. Anal ring terminal, with elongate-oval cells and 6 long setae. Ostioles present. Circulus present. Eyes present. Body setae short and sparse. Trilocular pores present. Tubular ducts present, each heavily sclerotized and slightly dome-shaped at inner end with, apparently, a septum. Bitubular and tritubular cerores absent.
COMMENTS. The only distinguishing characters sepa- rating this monotypic genus from Rhizoecus is the unusual type of duct. The genus, known from Califor- nia, is accepted by most workers on scale insects but further related species may help to clarify its position.
Rhizoecus Kiinckel d’ Herculais
Rhizoecus Kinckel d’Herculais. 1878: 163. Type spe- cies Rhizoecus falcifer Kiinckel d’Herculais, by monotypy.
Ripersiella Tinsley, in Cockerell 1899:278. Type spe- cies Ripersia rumicis Maskell, by subsequent designation of Cockerell, 1901: 165. Synonymised by Hambleton, 1974: 147.
Pararhizoecus Goux, 1941: 197, as a subgenus of Rhizoecus. Type species Rhizoecus (Pararhizoecus) petiti Goux, by original designation and monotypy.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 29
Synonymised with Ripersiella by Morrison & Morrison, 1966: 147, and with Rhizoecus by Ben- Dov, 1994: 286.
Morrisonella Hambleton, 1946a: 16. Type species Morrisonella poensis Hambleton, by original des- ignation. Homonym of Morrisonella Bartsch, 1920. Synonymised by Ferris, 1953: 426.
Radicoccus Hambleton, 1946a: 47. Williams & Granara de Willink, 1992: 492. Type species Rhizoecus globosus James, by original designa- tion. Syn. nov.
Coccidella Hambleton, 1946b: 177. Replacement name for Morrisonella Hambleton. Synonymised by Ferris, 1953: 426.
Neorhizoecus Hambleton, 1946a: 40. Type species Rhizoecus coffeae Laing, by original designation. Synonymised by Lindinger, 1957: 550.
DIAGNOSIS. Body normally small, elongate to ro- tund, anal lobes usually not developed, their positions each usually with 3 long anal lobe setae, or more rarely, with numerous setae. Antennae usually placed close together, short, strongly geniculate, each with 5 or 6 segments, these often wider than long, terminal segment tapering, almost triangular; with sensory falcate setae well developed on terminal and penulti- mate segments, these in addition to normal flagellate setae. Labium elongate, longer than wide. Anal ring with 6 setae, these usually flagellate or occasionally clavate; with fairly large elongate to triangular cells that can easily be counted to distinguish between species. Legs normally well developed; tarsus taper- ing to a long slender claw with short setose or dilated digitules, sometimes about as long as claw; without translucent pores but occasionally with large vacu- olate pores; setae on inner edges of tibia and tarsus often elongate spine-like. Eyes present or absent. Cir- culus present or absent, when present usually truncate conical, distal surface often reticulate or faveolate, usually situated on abdominal segment II within bor- ders of segment; sometimes as many as 6 circuli present distributed singly among other abdominal seg- ments and occasionally present on metathorax. Ostioles present, placed well on dorsum, sometimes reduced to posterior pair only and occasionally barely percepti- ble. Frons often with sclerotized cephalic plate.
Body setae usually short and flagellate, often abun- dant, rarely very sparse. Bilocular or trilocular cerores present, never both types present together, rarely ab- sent altogether. Trilocular pores present, often abundant, sometimes few. Tubular ducts present or absent, when present usually minute with parallel or subparallel sides. Multilocular disc pores present or absent. Medioventral pore groups sometimes present on abdomen. Mushroom bodies occasionally present on dorsum and venter.
COMMENTS. The above description should be ad-
equate to separate the genus from all others discussed here. The tritubular and bitubular cerores are unique to the genus and some attempts have been made to distin- guish between Rhizoecus with tritubular cerores and Ripersiella with bitubular cerores but there are many variations of the cerores. Sometimes the tubes of the bitubular cerores are so appressed as to appear unitubular and often their true structure can only be determined by the use of oil immersion. Besides, R. andensis (Hambleton) andR. colombiensis Hambleton, both South American species, and R. cobelopus Williams from Australia, are without either type but appear to be congeneric in other characters.
Emphasis on the presence or absence of eyes, the length and type of the claw digitules, and the shape of the bitubular and tritubular cerores to distinguish among genera has been challenged by Takagi & Kawai (1971) who commented on the variability of these characters.
The position of Radicoccus needs special mention. Hambleton (1946a) erected this genus for five species, each with a fairly stout body, the legs and antennae, although well developed, are minute for the size of body, and body setae are sparse. The species show widely diverse characters but these fall within the present concepts of Rhizoecus. These five species and another transferred since, are discussed as follows:
Rhizoecus globosus James. This African species, was illustrated by De Lotto (1957) and, considering the poor material on which the illustration was based, the illustration is fairly comprehensive. Some further notes are now added although a study of fresh material is needed. The species possesses 3 circuli and large tritubular cerores of two distinct sizes, each with short stout ducts giving the whole ceroris a flat appearance. Anterior and posterior ostioles are present although they are small and almost imperceptible.
Rhizoecus incrassatus James. This African species is similar to R. globosus, differing in possessing only 2 circuli and tritubular cerores of one size. Anterior and posterior ostioles are present. The species was illus- trated by De Lotto (1957).
Rhizoecus kelloggi Ehrhorn & Cockerell.This North American species has sparse body setae, lacks bitubular or tritubular cerores but possesses distinctive tubular ducts, each with the orifice elliptical; nevertheless, the sides of each duct are parallel. Ostioles are present although they are represented by mere slits (D.R. Miller, personal communication). There are no good reasons to exclude this species from Rhizoecus.
Radicoccus hawaiiensis Hambleton. It was shown by Beardsley (1966) that this species possesses all the characters of Rhizoecus to which it was transferred.
Rhizoecus poltavae Laing. For some reason this species, described from Ukraine, has always been regarded as distinct in lacking bitubular or tritubular cerores but bitubular cerores are plainly discernible in the original material, a character noted by Matile-
30
Ferrero (1976) who included the species in Ripersiella. The species is here retained in Rhizoecus.
Rhizoecus cocois Williams. Although this species,
described from India, was transferred to Radicoccus by Tang (1992) because of the globular body and paucity of body setae, the action was unnecessary. The species 1s related toR. globosus and R. incrassatus and should be retained in Rhizoecus.
All the species, therefore, that have been assigned to
Radicoccus can be comfortably included in Rhizoecus and the name Radicoccus, as listed by Ben-Dov (1994), is here synonymised with Rhizoecus.
Key to Genera of the Subfamily Rhizoecinae (Adult Females)
Anal ring with protuberances; anal ring setae short, much shorter than diameter of anal ring, numbering 2—5, un- evenly distributed .................. Pseudorhizoecus (p. 28)
Anal ring without protuberances, anal ring setae con- spicuous, usually at least as long as diameter of anal ring or almost as long, numbering 6-16, distributed evenly on ring or detached from ring and lying in an arc anterior to ring (sometimes anal ring masked by heavily sclerotized anal Lobes) bere vescas Sins sonasneanna cath oun tedeepetoeeeeaeveraneedis 2
Anal ring distinctly ventral in position, with a single row of elongate cells and thick spine-like anal ring setae around posterior edge (actually anterior edge if trans- HETHRSLS | (HO) GORA WHOM) oo eecocosecnceere Leptorhizoecus (p. 27)
Anal ring always dorsal in position, anal ring setae not spine-like, at least the 2 anterior pairs slender. Cells PLESCME OL ASEM ti A reeenese ene ceee cee acento eee 3)
Anal ring with large elongate to triangular cells, anal ring setae arising from surface of anal ring. Body rotund, elongate-oval or weakly pyriform, never with cephalothorax strongly dilated. Circulus, if present, never cupped in centre. Ostioles present or absent. Tubular GuctsipresentiomabSscm tases -eeee ee eee eee ere eee 4
Anal ring simple, without elongate cells, anal ring setae detached from ring or at edge of ring, usually lying in an are anterior to ring, sometimes difficult to distinguish
D.J. WILLIAMS sosneasptez row. Poh Bo, ee, Capitisetella (p. 26)
Anal lobes well developed, protruding well beyond pos- terior end of body, each lobe terminating in a stout SPINE -Wike\SCtal a wesmnee eee Geococcus (p. 26)
Anal lobes barely perceptible or not developed. Any setae on anal lobes or on normal positions of anal lobes, flagellate sro tas 1m SKC se cece reese ee ners ae tece eee 7
Body with conspicuous dome-shaped tubular ducts, each heavily sclerotized. Bitubular and tritubular cerores al- Way StaDSEM ities --s eee oe ene Pygmaeococcus (p. 28)
Body without conspicuous dome-shaped tubular ducts; if tubular ducts are present they are minute with parallel or subparallel sides. Bitubular or tritubular cerores often present. Body setae usually abundant, rarely sparse ..... $9. Se CE Lee. | ee i eee ee ae Rhizoecus (p. 28)
Anal lobes protruding, rounded, each with numerous flagellate setae. Antennae each with 4 or 5 segments. Anal opening situated at base of anal lobes. Trilocular pores present or absent. Circuli present .................2:000 AS A ee RA ee Re Neochavesia (p. 27)
Anal lobes not protruding, posterior end of body rounded, each lobe often with long setae, sometimes almost as long as body. Circuli present or absent. Antennae each with 1-4 segments. Trilocular pores always absent.....9
Antennae short, each with 24 segments; any articulation between first and second segments poorly developed. Claws normally long and slender except in one species with stout claws. Minute setae abundant, always present on dorsum and venter of abdomen, most on abdomen sometimes replaced with short sickle-shaped setae. Cla- vate sensory setae present or absent on body Eumyrmococcus (p. 8)
Antennae conspicuously long, always with 4 segments, almost as long as body, the first and second segments with well developed articulation, the proximal end of second segment with small points which fit into grooves at distal end of first segment. Claws long and slender. Minute setae abundant on dorsum of abdomen only, absent from venter of abdomen; ventral setae on abdo- men mostly long and stout. If sickle-shaped setae are present they form small groups on thorax only. Clavate
from dorsal setae of abdominal segment VIII. sensory setae always absent........... Xenococcus (p. 22) Cephalothorax dilated, abdomen narrow, tapering (in E. neoguineensis, the cephalothorax is dilated but the abdo- men widens before tapering). Circulus, if present, cupped in centre, even if only shallowly. Ostioles always absent. Tubularductsyalway S| abSentt cue .c--eesse.eeecveeseatense eee 8 RAT CTR SETS Tes 4 All body setae knobbed. Antennae each with 3 or 4 WITH DISCUSSION OF SESMENtS LN ae ee ee ee: 5 SPECIFICITY OF THE MEALYBUG- — All body setae flagellate. Antennae each with 5 or 6 ANT ASSOCIATIONS
SEG ITEM EG Soc. spac nec tat base pce vse teres aoe ee ote eRe 6
5 Antennae each with 4 segments. Multilocular disc pores present. Anal ring with 8—16 short capitate setae .......... an Sas a tn eine ein Brevicoccus (p. 26)
Eumyrmococcus corinthiacus, Greece, with Acropyga (Rhizomyrma) paleartica, with queen in flight. Eumyrmococcus falciculosus, Sumatra with ? ant, pri-
mary forest litter, Hevea soil. Eumyrmococcus kolombangarae, Solomon Islands,
— Antennae each with 3 segments. Multilocular disc pores absent. Anal ring with 6 pointed setae ..............0000000
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 31
withAcropyga (Rhizomyrma) lauta in log, queen in flight.
Eumyrmococcus kruiensis, Sumatra, with ? ant, in soil.
Eumyrmococcus kusiacus, Solomon Islands, with Acropyga (Rhizomyrma) lauta in log, queen in flight.
Eumyrmococcus lamondicus, Australia, withAcropyga sp. in rotting wood.
Eumyrmococcus lanuginosus, Sumatra, with ‘ants’.
Eumyrmococcus maninjauensis, Sumatra, with ? ants, in soil.
Eumyrmococcus neoguineensis, Papua New Guinea, with Acropyga (Atopodon) ambigua, in rotting log.
Eumyrmococcus nipponensis, Japan, with Acropyga (Atopodon) nipponensis, on rootlets of plants or in rotten wood.
Eumyrmococcus recalvus, Australia, with Acropyga sp.
Eumyrmococcus sarawakensis, Sarawak, with Acropyga sp.
Eumyrmococcus scorpioides, South Africa, with Acropyga (Malacomyrma) arnoldi, in nests and mandibles.
Eumyrmococcus smithii, SE. Asia, with Acropyga (Rhizomyrma) sauteri, in nests.
Eumyrmococcus sulawesicus, Sulawesi, withAcropyga sp., rain forest.
Eumyrmococcus taylori, Australia, Acropyga sp., in nests.
Eumyrmococcus sp., Papua New Guinea, with Acropyga (Rhizomyrma) oceanica, in log.
Eumyrmococcus sp., Papua New Guinea, with Acropyga sp., in rainforest.
Xenococcus acropygae, Southern Asia, Australasia, with Acropyga (Acropyga) acutiventris, A. (Atopodon) ambigua, A. (Rhizomyrma) lauta, in nests and mandibles of flying queens.
Xenococcus annandalei, India, with Acropyga (Acropyga) acutiventris, in nests and with flying queens.
Neochavesia caldasiae, Colombia, Trinidad, with Acropyga (Rhizomyrma) robae.
Neochavesia eversi, Colombia, Panama, withAcropyga (Rhizomyrma) kathryna.
Neochavesia trinidadensis, Colombia, Trinidad, with Acropyga sp.
Neochavesia weberi, Guyana, with Acropyga (Rhizomyrma) paludis.
Capitisetella migrans, Colombia, Guyana, Surinam, Trinidad, with Acropyga (Rhizomyrma) rutgersi.
Pseudorhizoecus proximus, Colombia, Costa Rica, Ecuador, Guatemala, Surinam, with Acropyga (Rhizomyrma) rutgersi, A. (Rhizomyrma) para- mibensis.
Rhizoecus coffeae, Brazil, Costa Rica, Surinam, with Acropyga (Rhizomyrma) paramibensis.
Rhizoecus caladii, Brazil, Colombia, Guyana, Surinam, with Acropyga (Rhizomyrma) paramibensis.
Rhizoecus moruliferus (= R. flalcifer), Surinam, Trini- dad, with Acropyga (Rhizomyrma) paramibensis.
Geococcus coffeae, South America, with Acropyga (Rhizomyrma) paramibensis.
Where possible the ant species are listed in the four subgenera of Acropyga presented by Bolton (1995a). Other Rhizoecinae associated with Acropyga species in South America have been recorded mentioning mealybug genera only. It is not certain if voucher specimens were kept and at present the species cannot be identified.
There is no doubt that Xenococcus acropygae and X. annandalei are always attented by Acropyga (Acropyga) acutiventris and the mealybug may be found eventually wherever the ant is located. The symbiotic association suggests that neither ant nor mealybug can live without the other. Taylor (1992) discusses the distribution of the ant in more detail with special reference toAustralia. Records of A. (Atopodon) ambigua in Papua New Guinea and of A. (Rhizomyrma) lauta in the Solomon Islands, associated with X. acropygae, cannot be taken too seriously because they are listed from spirit material only. The type locality of A. acutiventris is Sri Lanka but so far no species of Xenococcus have been reported from there despite extensive collecting of mealybugs by E.E. Green at the beginning of this century.
A much better picture of ant-mealybug specificity would have emerged if some of the ants had been identified and recorded. Present records indicate that each species of Eumyrmococcus may be attended by a different ant species. The type species of Eumyrmo- coccus, E. smithii, has an apparent wider distribution than any other in the genus and it is always attended by A. (Rhizomyrma) sauteri. E. kolombangarae and E. kusiacus, two closely-related species, are attended by A. (Rhizomyrma) lauta in the Solomon Islands but the mealybugs are known from only a few specimens. E. neoguineensis is attended by A. (Atopodon) ambigua in Papua New Guinea and an interesting species of Eumyrmococcus, known from immatures only, is as- sociated with A. (Rhizomyrma) oceanica. these mealybugs in the Solomon Islands and Papua New Guinea live close to colonies of A. acutiventris attend- ing X. acropygae.
In Europe, E. corinthiacus is apparently associated withA. (Rhizomyrma) paleartica and E. scorpioides in SouthAfrica is attended by A. (Malacomyrma) arnoldi.
Central and South American mealybug species in other genera are always associated with Rhizomyrma, the only subgenus in the area, and normally each mealybug species appears to be specific to a particular ant species.
The wide gap in records of Eumyrmococcus be-
Sy
tween Europe and South Africa in the west and south- ern Asia and Australasia in the east may be due to lack of collecting. Another possibility could be that spe- cies of Eumyrmococcus have been collected there but, at present, are lying in ant vials waiting to be sorted or are present in soil sample material extracted by stu- dents of Acari or Collembola.
There must be instances when ant species overlap and it seems difficult to believe that any species of Acropyga would reject a strange species of Eumyrmococcus if it were offered. Geococcus coffeae is readily attended by A. (Rhizomyrma) paramaribensis in Brazil although the mealybug origi- nated in southern Asia.
These listed associations probably represent only a fraction of similar Acropyga-mealybug associations throughout southern Asia and Australasia. The asso- ciations may be widespread and just await further study.
The mealybug—ant system offers itself as an excel- lent future candidate for detailed coevolutionary studies based on independent cladistic analyses of the partner groups.
REFERENCES
Balachowsky, A.S. 1957. Sur un nouveau genre aberrant de cochenille radicole myrmécophile nuisible au caféier en Colombie. Revue de Pathologie Végétale et d'EntomologieAgricole de France 36: 157-164.
Beardsley, J.W. 1962. Descriptions and notes on male mealybugs (Homoptera: Pseudococcidae). Proceedings of the Hawaiian En- tomological Society 18: 81-98.
Beardsley, J.W. 1966. Hypogeic mealybugs of the Hawaiian Islands (Homoptera: Pseudococcidae). Proceedings of the Hawaiian En- tomological Society 19; 151-155.
Beardsley, J.W. 1970. Three new species of Chavesia Balachowsky
from tropical America (Homoptera: Coccoidea). Proceedings of
the Hawaiian Entomological Society 20: 509-520.
Ben-Doy, Y. 1994. A systematic catalogue of the mealybugs of the world (Insecta: Homoptera: Coccoidea: Pseudococcidae and Putoidae) with data on geographical distribution, host plants, biology and economic importance. 686 pp. Intercept Ltd, Andover.
Bolton, B. 1995a. A new general catalogue of the ants of the world. 504 pp. University Press, Cambridge, Mass.; London.
Bolton, B. 1995b. A taxonomic and zoogeographical census of the extant ant taxa (Hymenoptera: Formicidae). Journal of Natural History 29: 1037-1056.
Buchner, P. 1957. Endosymbiosestudien an Schildléusen V. die Gattung Rastrococcus Ferris (Ceroputo Sule). Zeitschrift fiir Morphologie und Okologie der Tiere 46: 111-148.
Buchner, P. 1965. Endosymbiosis of animals with plant microorgan- isms. Revised English version. 909 pp. Interscience Publishers, New York, London, Sydney.
Biinzli, G.H. 1935, Untersuchungen tiber coccidophile Ameisen aus den Kaffeefeldern von Surinam. Mitteilungen der Schweizerischen Entomologischen Gesellschaft 16: 455-593.
Buschinger, A., Heinze, J., Jessen, K., Douwes, P. & Winter, U. 1987. First European record of a queen ant carrying a mealybug during her mating flight. Naturwissenschaften 74: 130-140.
Cockerell, T.D.A. 1899. Tables for the determination of the genera of Coccidae. The Canadian Entomologist 31: 237-279.
D.J. WILLIAMS
Cockerell, T.D.A. 1901. New and little-known Coccidae. 1. Ripersiella and Ceroputo. Proceedings of the Biological Society of Washington 14: 165-167.
Cox, J.M. & Pearce, M.J. 1983. Wax produced by dermal pores in three species of mealybug (Homoptera: Pseudococcidae). Interna- tional Journal of Insect Morphology and Embryology 12: 235-248.
De Lotto, G. 1957. The Pseudococcidae (Hom.: Coccoidea) described by H.C. James from East Africa. Bulletin of the British Museum (Natural History) (Entomology) 5: 185-232.
De Lotto, G. 1977. On some African mealybugs (Homoptera: Coccoidea: Pseudococcidae). Journal of the Entomological Society of Southern Africa 40: 13-36.
Ferriere, C. 1956. Encyrtides parasites de cochenilles sur Graminées. Bollettino Laboratorio di Zoologia Generale eAgraria della Facolta Agraria in Portici 33: 350-364.
Ferris, G.F. 1953. Atlas of the scale insects of North America. Volume VI. The Pseudococcidae (Part II). pp. 279-506. Stanford University Press, California.
Ferris, G.F. & Murdock, G.E. 1936. Contributions to the knowledge of the Coccoidea (Homoptera). III. On certain dermal structures of the Pseudococcidae. Microentomology 1: 115-122.
Flanders, S.E. 1957. The complete interdependence of an ant and a coccid. Ecology 38: 535-536.
Goux, L. 1941. Description d’un Rhizoecus nouveau et de sa larve néonate. Bulletin du Musée d'Histoire Naturelle de Marseille 1: 197-203.
Green, E.E. 1902. Three new genera of Coccidae from Ceylon. Entomologist’s Monthly Magazine 38: 260-263.
Green, E.E. 1933. Notes on some Coccidae from Surinam, Dutch Guiana, with descriptions of new species. Stylops 2: 49-58.
Gullan, P.J. & Kosztarab, M. 1997. Adaptations in scale insects. Annual Review of Entomology 42: 23-50.
Hambleton, E.J. 1946a. Studies of hypogeic mealybugs. Revista de Entomologia 17: \-77.
Hambleton, E.J. 1946b. A new name for a mealybug. Proceedings of the Biological Society of Washington 59: 17.
Hambleton, E.J. 1974. Three new species of Rhizoecus (Homoptera: Pseudococcidae) from New Zealand, with notes and redescriptions of others. New Zealand Journal of Zoology 1: 147-158.
Hambleton, E.J. 1976. A revision of the New World mealybugs of the genus Rhizoecus (Homoptera: Pseudococcidae). Technical Bulletin United States Department of Agriculture 1522: 1-88.
Hambleton, E.J. 1977. A review of Pseudorhizoecus Green, with a description of a related genus (Homoptera: Pseudococcidae). Jour- nal of the Washington Academy of Sciences 67: 38-41.
Hambleton, E.J. 1979. New information on the Rhizoecus of Florida including descriptions of four new species. The Florida Entomolo- gist 62: 140-149.
Koteja, J. 1974a. Comparative studies on the labium in the Coccoidea (Homoptera). Zeszyty Naukowe Akademii Rolniczej w Krakowie 27: 1-162.
Koteja, J. 1974b. On the phylogeny and classification of the scale insects (Homoptera, Coccinea) (discussion based on the morphol- ogy of the mouthparts). Acta Zoologica Cracoviensia 19: 267-325.
Koteja, J. 1984. Why the scale insects (Homoptera, Coccinea) are unusual. Verhandlungen des Zehnten Internationalen Symposiums tiber Entomofaunistik Mitteleuropas (SIEEC) 15.-20. August 1983 Budapest pp. 325-327.
Koteja, J. 1985. Essay on the prehistory of the scale insects (Homoptera, Coccinea). Annales Zoologici 38: 461-503.
Kiinckel d’Herculais, J. 1878. Histoire da la cochenille vivant sur les racines de Palmiers de la section des Seaforthia. Exposé des caractéres du genre Rhizoecus. Annales de la Société Entomologique de France 8: 161-164.
Lindinger, L. 1957. Ein weiter Beitrag zur Synonymie der Cocciden. Beitrdge zur Entomologie 7: 543-553.
Lloyd, D.C. & Martini, E. 1957. A note on the circulus as an adhesive organ in some Pseudococcidae. Canadian Entomologist 89: 46-48.
MacGillivray, A.D. 1921. The Coccidae. Tables for the identification of the subfamilies and some of the more important genera and
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 33
species, together with discussions of their anatomy and life history. vi, 502 pp. Scarab Company, Urbana, Illinois.
Matile-Ferrero, D. 1976. La faune terrestre de I’ [le de Sainte-Héléne. 7. Coccoidea. Annales Musée Royal de l'Afrique Centrale, Tervuren. Série in 8°. Sciences Zoologiques 215: 292-318.
McKenzie, H.L. 1960. Taxonomic study of California mealybugs with descriptions of new species (Homoptera: Coccoidea: Pseudococcidae). Hilgardia 29: 681-770.
Miller, D.R. & McKenzie, H.L. 1971. Sixth taxonomic study of North American mealybugs, with additional species from South America (Homoptera: Coccoidea: Pseudococcidae). Hilgardia 40: 565-602.
Miller, D.R. & Williams, D.J. 1995. Systematic revision of the family Micrococcidae (Homoptera: Coccoidea), with a discussion of its relationships, and a description of a gynandromorph. Bollettino del Laboratorio di Entomologia Agraria ‘Filippo Silvestri’, Portici [1993] 50: 199-247.
Morrison, H. & Morrison, E.R. 1966. An annotated list of generic names of the scale insects (Homoptera: Coccoidea). Miscellaneous Publications United States Department ofAgriculture 1015: 1-206.
Noyes. J.S. & Hayat, M. 1994. Oriental mealybug parasitoids of the Anagyrini (Hymenoptera: Encyrtidae) with a world review of Encyrtidae used in classical biological control and an index of encyrtid parasitoids of mealybugs (Homoptera: Pseudococcidae). 554 pp. CAB International, Wallingford; The Natural History Museum, London.
Pesson, P. 1939. Une glande exsertile ventrale chez les Coccides. Bulletin Société Zoologique de France 64; 144-152.
Prins, A.J. 1982. Review of Anoplolepis with reference to male genitalia, and notes on Acropyga (Hymenoptera, Formicidae). Annals of the South African Museum 89; 215-247.
Roepke, W. 1930. Ueber einen merkwiirdigen Fall von *Myrmekophilie’, bei einerAmeise(Cladomyrma sp.?) auf Sumatra beobachtet. Miscellanea Zoologica Sumatrana 45: |-3.
Schmutterer, H. 1952. Die Okologie der Cocciden (Homoptera, Coccoidea) Frankens |. Abschnitt). Zeitschrift fiir Angewandte Entomologie 33: 369-420.
Silvestri, F. 1924. A new myrmecophilous genus of Coccidae (Hemi- ptera) from India. Records of the Indian Museum 26: 311-315. Silvestri, F. 1926. Descrizione di un novo genere di Coccidae (Hemi- ptera) mirmecofilo della Cina. Bollettino del Laboratorio di Zoologia Generale e Agraria della R. Scuola Superiore
d’Agricoltura in Portici [1925] 18; 271-275.
Silvestri, F. 1927. Descriptions of two species of myrmecophilous Coccidae (Hemiptera Insecta). China Journal 7: 253-254.
Takagi, S. & Kawai, S. 1971. Two new hypogeic mealybugs of Rhizoecus from Japan (Homoptera: Coccoidea). Kontyfi, Tokyo 39: 373-378.
Takahashi, R. 1934. Observations on the Coccidae of Formosa, Part IV. Report of the Government Research Institute Department of Agriculture Formosa 63: \-38.
Tang, ET. 1992. The Pseudococcidae of China. 767 pp. Chinese Agricultural Science Technology Press, Beijing.
Taylor, R.W. 1992. Nomenclature and distribution of some Austral- ian and New Guinean ants of the subfamily Formicinae
(Hymenoptera: Formicidae). Journal of the Australian Entomo- logical Society 31: 57-69.
Taylor, R.W, & Brown, D.R. 1985. Formicoidea. pp. 1-149. Jn Taylor, R.W. & Brown, D.R. (Eds) Zoological Catalogue of Aus- tralia Volume 2 Hymenoptera: Formicoidea, Vespoidea and Sphecoidea. 381 pp. Bureau of Flora and Fauna, Canberra.
Terayama, M. 1985. Two new species of the genus Acropyga (Hy- menoptera, Formicidae) from Taiwan and Japan. Kontyi, Tokyo 53: 284-289.
Terayama, M. 1986. A new species of anomalous ant-attended mealybug genus Eumyrmococcus (Homoptera, Pseudococcidae) from Japan. Kontyii, Tokyo 54: 509-512.
Terayama, M. 1988. Some taxonomical and biological notes on the myrmecophilous mealybug genus Ewmnyrmococcus (Homoptera: Pseudococcidae) Rostria 39: 643-648.
Tremblay, E. 1977. Advances in endosymbiont studies in Coccoidea. Research Division Bulletin, Virginia Polytechnic Institute and State University 127: 23-33.
Tremblay, E. 1989. Coccoidea endocytobiosis, pp. 145-173. In Schwemmler, W. & Gassner, G. (Eds). Insect endocytobiosis, mor- phology, physiology, genetics, evolution. 272 pp. CRC Press, Boca Raton, Florida.
Williams, D.J. 1969a. The family-group names of the scale insects (Hemiptera: Coccoidea). Bulletin of the British Museum (Natural History) (Entomology) 23: 315-341.
Williams, D.J. 1969b. A revision of the genus Geococcus Green (Homoptera, Coccoidea, Pseudococcidae). Bulletin of Entomo- logical Research 59: 505-517.
Williams, D.J. 1970. The mealybugs (Homoptera. Coccoidea, Pseudococcidae) of sugar-cane, rice and sorghum. Bulletin of Entomological Research 60: 109-188.
Williams, D.J. 1978. The anomalous ant-attended mealybugs (Homoptera: Pseudococcidae) of south-east Asia. Bulletin of the British Museum (Natural History) (Entomology) 37: 1-72.
Williams, D.J. 1985. Australian mealybugs. vii, 431 pp. British Museum (Natural History), London.
Williams, D.J. 1988. Unusual instars in Xenococcus annandalei Silvestri (Homoptera: Pseudococcidae). Bollettino del Laboratorio di Entomologia Agraria ‘Filippo Silvestri’, Portici [1986] 43: 73- oie
Williams, D.J. 1993. A new species of mealybug from Greece, the first from Europe belonging to the ant-attended genus Eumyrmococcus Silvestri (Hemiptera: Coccoidea: Pseudococcidae) Entomologist’s Gazette 44: 216-220.
Williams, D.J. 1996. Four related species of root mealybugs of the genus Rhizoecus from east and southeast Asia of importance at quarantine inspection (Hemiptera: Coccoidea: Pseudococcidae). Journal of Natural History 30: 1391-1403.
Williams, D.J. & Granara de Willink, M.C. 1992. Mealybugs of Central and South America, 635 pp. C.A.B. International, Wallingford.
Williams, D.J. & Watson, G.W. 1988. The scale insects of the tropical South Pacific region, Part 2, The mealybugs (Pseudococcidae). 260 pp. C.A.B. International, Wallingford.
34 D.J. WILLIAMS
Fig. 1 Eumyrmococcus taylori sp. n. A. Male pupa enclosing adult male. B. Female pupa enclosing adult female.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI 35
\ \ WAAL Wary. SEQ NAN Ue
ii i AG MK fan HiAAT Ta j - tii in vif titli uy i th i a
pany
a Nit sae iit TAA)
Fig. 2} Ewmyrmococcus corinthiacus Williams. Adult female.
36 D.J. WILLIAMS
a
f /t FN Tay tre 2 OUR WNP UPA I IAS
VA re Ua
es ’ tr Dy PEN TASST NAAN = Be LOAN OSC ee a olia nat OATS Se e FONG MUNA ALA VINE. Tay OTS YN TT IGY ~ Bae fad) May tye iW WY AVIVA AQ EY yay! ee ie 5 PSA I AO A Aca A is ee oe in HIS NF Sy 3 - AG a ~ a a Zyl NA NYAS S AAR TATA GAR AMY NIN ys Za GINTAMA APN AAA DAES - WER AL DIA Ma REE Le HARTANNIS TAUNTS GUEANY Mya ty oi Zz —_ = an # : f WYP Zan a Z CZ yp SVP ey ag Pe Z PNR TAR IAAL PA RRA TARR WAN) = 4 PUCK M ROR NAN UAV AAT /\\ rie RAN AIAN PITTING ana! Baas \ ' LA] NCA 4 iaNvy 2) Pe ae Sia
Fig. 3 Eumyrmococcus falciculosus sp. n. Adult female.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI Sif
La) LLL
i;
UG ZEEE LZ ZL Zz Zi io
\\
iy \ “
hit (nt wy ih Yy Bi. iy bas Fg atl 7 I Ny IAS EAA ZR fs gy TiN ANA ve Ye Vn ANY AY WN Y Ayn tis NN A if Lyin Lae i |
AN — \ W144 \\ A ea AY \\ At Mi \ iN Hat A ne AN AN NY Me "ips \ HIN UU NN ANY i Nh j eT itl AN a Hy NNN i Nw \\\ WAN AN NN i i INN BO \\ \
WANS i MAN WAN S Wi \ x i W\ | QO Ti NY RR AN MN tt Wy (Ary AW . ee A AN “r “ Wy ih At WWQWie ANY AK AN NN AK ANG Nf 1\! WY AW A NK AYN my AN AS 1A \\ its W WIQWAY \ \ \ caw NX \X SA\ 1 * \ 1)
WY
NX ANY
_ sc Ve \ AR \\\ \
WN tie Nea
on ih \! ANS WAX : wn
AN
=
S55
—S
=S
! HP f si HA nn Wi We WN Me hi
i! i at \ 2 heigl in ~
Fig. 4 Eumyrmococcus kolombangarae sp. n. Adult female.
38
Fig. 5 Eumyrmococcus kruiensis sp. n. Adult female.
D.J. WILLIAMS
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
7
bis) WIZZ
i iy
Yt
ZA lh HU Wi, Mi j My! eit
Lf, oe mane
>
\ \ a AN
Ni ita a WY : i ve IN
\ vit ‘i AN NW Wy a
\\\ aT AN nN \\ A S\\\ ‘ \
ANN \\\\ AN YS \ A NN AAN WY: WAN “i AY WN ANY Nit AN aN \\ \ WN \\ AN NAS in LN AWN is NY \\\ AN WANG iN AN AA HAN WN IAN
KAY AN NOK ANN Ay) AY \ \\ AN va Y \\\y \\y' \’ Wy wy \ \\\
S\N \ Wh Nh yh Ww AAS aN NW WHIRL NS \ WK A \i \\ AY Sh
AY Wi nh! ‘ LA AAW
\ ANY ANA Wh \ WHA
AN WW YY VAAN WN ‘ NN AY Nl AY ayy
\\ \ AAAS DANY \ WN Ny \ Wath Wea A ANY NY WY
Wt
\ NORA) \\ \ \ vA SU A Wh 4 \ YS We
ANAS |
4 wt ih ln -[- ‘ WW\ \ \\ iy! Til Vi I (| ' Ai \ en !
wy aN
ARR ANE NS HWW
\ hei HIN f / areas AN Ly Cet ily val ma Lod it I\Y AC NA 4 “ i uF me will iN) Cee, ae VEN qf li} ~ AE IT A Liaise eo it V, en Ny \\ \ \ aN Rites HVA ze UMN tile. iy, e eae oa os ce a Vax ee RRO a IW
Fig. 6 Eumyrmococcus kusiacus sp. n. Adult female.
40
is
ye cietc bts tif, GEG I LENG ohn Leathe Wh ih j CAAA 7 AAA > Lebo! Uh IW NMG Gls VAAL A LR HL fp eh NLU 1 G1 OGL iY pl l4g WG)
\ \
Fig. 7 Eumyrmococcus lamondicus sp. n. Adult female.
Y,
S VAN A
QQ)
NY
“
D.J. WILLIAMS
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
Pp i Pac AS} k len Sr i f Goce x7 ile oF Ad My re a t Peph ep ft Li Rat a) . ATR ne aM aa NG inte ji ra aviay negeart U Dl Aare) AW aire as eee ho MM Ece RAAT Pe a i pee i es ) w/'7 VU NS Pj ) - N
Vii Lhe Ris rip yy.
thitlyrs\e ant <i) MFP yy
: NI
- RRP FO ape hp s + \
r ade at tite int iat; i) ip GAT ie a) i; a ih hebjeleli “IR BOO. Pg Me eel ST Tat in| p ey ad ey }
en ee a aly eg eyy } i | PIP ify iy iil" ay > afcie I
Pay wish al |
ahr pile Lie i TA AAA ee Hh Hi
Linas
TY
fe if ae rlipcli Vy
| Lia, ; ZZ
a ra lak y
DJ.W.
Fig. 8 Eumyrmococcus lanuginosus sp. n. Adult female.
4]
42
D.J. WILLIAMS
Fig.9 Eumyrmococcus maninjauensis sp. n. Adult female.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
4 fam , 1
Pe UA) 1 ee ANN ater PLT (hays
if witen
aaa AT
bi v Han nf anh Hh Na \ i Ha)
Mf MA Ht sath ETAL ASR RON itt
NAN | AAT aye 1) Lie or reat te ir! NEN ars me A BEA AHA a AO nif ih ANON CAAA q ‘ ‘ ul Mat i RUNS NORA TORE
ie A SO UR
Lettnpientess it AN set ek ait Wi vi ne i ie Kp fapbuetp ey wt \s tit f NUN eh Get
yi CAPITA
FANT { Pa ON aA ENR Hy CAR wit we rat fi
ON) " yynieyd , % ante HARA MAY RH ROR
ity i AAAS ARA ARR AAAS SAMA anette OT fay
Mn iN ayaa at ert nhl Wt nti} H mit a why
Ra
a,
irae UMM pity | WM Aah gar nyt hwy ‘anna ey I
Pr IATL URAL ATA RTT R ORS DOR aaa re / fiji anny Wait ly cota
ALTON ea
, ii Lye \
na Ni
ut i f . AN Art Me ry nt 1
a a hth PRATT Mey ty Hy
wn na \
uM PMA Nye @
DANA iN iy Voy
i wi ant NN
y ia hy hy { Fit te me
Fig. 10 =Eumyrmococcus neoguineensis sp. n. Adult female.
44 D.J. WILLIAMS
Zi: i heey Pt
NA na
a Ki
ONG \ CaN
TaN Baw iP aN
shew
7 7 / a 4 7 7 Z ~ “= iw f \ - ~N “N Pa _ _ i ‘ XN N x . ~ ‘
D.J.W.
Fig. 11 9 Eumyrmococcus nipponensis Terayama. Adult female.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
(fitter y J \
MRAP LAG) t Fe ALOT oh
MTA ata (Vinny
NN aad
TARE ANH Rua
i It yl 1) i
V | Li}
HA EUAN EE |
Filip Hertha ye ahve
rye WV wernt
Nu HANS fl has ee =
\\ HTN j AANA | Ni wh AN F | a zg = ee eS TAN es itt
Fig. 12} Eumyrmococcus queenslandicus sp. n. Adult female.
45
46 D.J. WILLIAMS
C aS S SS . X a= — —— ~ = y inl mY ~ =a gL! Ui ~ i NTA + ih ~ ee iF /. ne ey x SS SSS 4 = ~ to WANG ; H = Vid )\ \ mgt th FY j u uy = Mf 1 Vi ‘ we ATI NN NH RA \ hat Front h AT Ml ee Mr tt Loe vir! wit! Tite Viti p gly, Nye iy ee USS = — rili' ~ < Ne i Nea ia EPA VEG a Mal hs S \ ai N Si atl tatiana Titi ~ Nau ~ S X = S ~ ~ TAWA f ANH RN Hifi WVtPSR 7 { i] Vr a _ Ss C - if ~ A Till ht Ss Sr Lily (yin ~ ~ Hy = PAI a & x Gre ~ a ; 3
Fig. 13. Eumyrmococcus recalvus sp. n. Adult female.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
Fig. 14 Eumyrmococcus sarawakensis sp. n. Adult female.
D.J. WILLIAMS
48
Fig. 15 Eumyrmococcus sarawakensis sp. n. Adult male.
49
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
CE, a
Z
/
VY
( ‘ \ - i ( a\j i
‘ al = i —o— ——_ ¢ iS : Jc L eo | =e | Sage ao ee '. NS ee Sse \ mee eee me < ae | OE SSSe= AS \ ey ‘
Fig. 16 Eumyrmococcus scorpioides (De Lotto). Adult female.
D.J. WILLIAMS
7: AH AM Mil
mit at
eS ks . AY NS
NN &
\\ me IAS AA \\
Was a
a A ‘\
S INT NAS \ AK hex
\ i\
enn i i
NY AW
sua NNER TE lh ed i uh Aan ttt) i inna al Aiea gar A Ty A SN etintatiny i
/ S CA Us
Fig. 17 Eumyrmococcus smithii Silvestri. Adult female.
Ft) y
bgt
A
ian] if
f j yy eee aa a aa } HEN ‘ li aes wy Pea
Fig. 18 Eumyrmococcus sulawesicus sp. n. Adult female.
all
52 D.J. WILLIAMS
it Le Zz Vikt E WV = - 2S = 3 S— wag a ee i [Sees 2 s y j & LD Vy 5, ri i‘ ae yh ut - | ’ 2 zZ Ze S S = x < < ~ ~ “N S < < < < x ~ S mS =
Fig. 19 Eumyrmococcus taylori sp. n. Adult female.
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
Fig. 20 Eumyrmococcus taylori sp. n. Second instar.
1 7 4 ¢
pe a/ W a Va
53
54
Fig.
21
Female pupa.
B. Male prepupa. C. Male pupa.
D.J. WILLIAMS
Fig. 22 Eumyrmococcus taylori sp. n. Adult male.
56
Leg apy
eens A iph fit
PAP TT
Fig. 23 Xenococcus acropygae sp. n. Adult female.
D.J. WILLIAMS
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
D.J.W.
de sp. n. First instar.
58
Fig. 25 Xenococcus acropygae sp. n. A. Female pupa. B. Male prepupa. C. Male pupa.
D.J. WILLIAMS
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
59
60 D.J. WILLIAMS
Ut PRiy ul anit
allt TU
me 7 Tyg Cl mnt
TULA Ve A
Fits alr
TMT ia Hiysthy 1 aut i K TANS it
a Aa i a ARAN) GIN ea aa TTL THY AA ALTA eH TTA NY,
f
Fig. 27 Xenococcus annandalei Silvestri. Adult female.
Fig. 28 Leptorhizoecus deharvengi sp. n. Adult female.
61
D.J. WILLIAMS
Mojany “LI SNIISAMDINS “7 O] HYNUs “FT CI saploidioos “7 p| SISUBYDMDADS “FT ¢] snajp2a4 “J TZ] snoipupjsuaanb “FZ || sisuauoddiu “7 (| SIsuaaUInsoau “FT 6 sisuanviuiupu “J g snsourgnup] “J 1, snoipuoun] ‘yg SnovISny “y ¢ sIsuaIndy “7 p apaDsunquojoy “FT ¢ snsojnoi20f “q 7 SNIDIYIUIAOD “7 |
NS ON
S
63
MEALYBUGS OF THE GENERA EUMYRMOCOCCUS SILVESTRI AND XENOCOCCUS SILVESTRI
‘sno2020uaX JO saidads Jo deul uonngmnsIq = O¢ “SIA
D.J. WILLIAMS
INDEX
Synonyms and misidentifications are in italics; main citations in bold
acropygae 4, 23, 31 albidus 4, 5 andensis 29 annandalei 23, 25, 31 annandalei 23 atopoporus 28
Brevicoccus 26, 30
caladii 4, 31
caldasiae 4, 27, 31 Capitisetella 26, 30 Chavesia 27 clavisetosus 26 cobelopus 29 Coccidella 29
cocois 30
coffeae Geococcus 4, 31 coffeae Rhizoecus 4, 29, 31 colombiensis 29 corinthiacus 9, 10, 30
deharvengi 27
Eumyrmococcus 8, 30, 31 eversi 4, 5, 31
falciculosus 9, 10, 30 falcifer 4,5, 28, 31
Geococcus 26, 30
globosus 29 hawaiiensis 29 incrassatus 29
kelloggi 29 kolombangarae 9, 11, 30 kruiensis 9, 11, 31 kusiacus 9, 12, 31
lamondicus 9, 12, 31 lanuginosus 9, 13, 31 Leptorhizoecus 27, 30
maninjauensis 9, 14, 31 migrans 4, 26, 31 Morrisonella 29 morrisoni 28 moruliferus 4, 31
Neochavesia 27, 30 neoguineensis 9, 14, 31 Neorhizoecus 29 nipponensis 9, 15, 31
Pararhizoecus 28 petiti 28
poensis 29 poltavae 29 Prorhizoecina 26
Prorhizoecus 28 proximus 4, 28, 31 Pseudorhizoecina 26 Pseudorhizoecus 28, 30 Pygmaeococcus 28, 30
queenslandicus 9, 16
Radicoccus 29 radicum 26 recalvus 10, 17, 31 Rhizoecina 26 Rhizoecinae 2, 26 Rhizoecini 2 Rhizoecus 28, 30 Ripersiella 28 rumicis 28
sarawakensis 9, 18, 31 scorpioides 9, 19, 31 smithii 9, 19, 31 sulawesicus 9, 20, 31
taylori 10, 20, 31 trinidadensis 4, 31
weber 31
Xenococcini 2 Xenococcus 22, 30
Bull. nat. Hist. Mus. Lond. (Ent.) 67(1): 65-78 Issued 25 June 1998
Monophyly of the dacetonine tribe- group and its component tribes (Hymenoptera: Formicidae)
BARRY BOLTON
Department of Entomology, The Natural History Museum, Cromwell Road, London SW7 SBD, UK (e-mail: bb@nhm.ac.uk)
CONTENTS MRO CLIN C HOM copes sso en snaps os pacndids<csieceen da sesteeecienaepohu2 cn ects Geegee tsa ae vnc tupeeehe: San comms Rate Uns oS 65 BRAUN COIN ONAL G US SCONY secs arses snsne ao dci a axcionsanvnanend sanadesusscdanavdasatus leas tadeevts dite sateite Sevas dup adnan stacyiasvaniaens 66 Monophvilyion thle acetone (lb 6 OUOMD cece ececxs ce ¥aeenagenscucheessievey n=aetlsue neunsasues, cbekvs esaes-.2a¥ 67 Syuapomorpoies Of the GacetomiMe Ee PTOUP! cores cxceecee- -tedewneetrencanvavaceeaech ten antace- esters <oree es 68 Some characters considered but not used at tribe-group rank ...........cccccceeceseeeeeseeseeseeseeeeeees 70 IVEOTLO PLLVLys Ot OEE ASICEN CULT eee te ee enter cscs cxcnes cvtuna hex tatar Farosteb a ueeaaeleenaseess ceanupsceteasare coders 70 FSDOMONPMIGS OL ULE EAS ICCLOLIU ee steeesaes.: sac: efutacectosntsnacenss sacrsayedasaaescts saeanosvaseacechteganseaness 70 Mono PIL y Ly OLDE DD ACSLOMIMI ere: eatetetet ctee acs n<o shades teeaunn snaxtrsstensenserceesta Tons wtncpscsecennteneeesenace 71 APOMONPRIES OM LDS) ACELOMINM cert ye.stetcsss.oncntvenosnctunavectcrondeebesstav ober steasttseioesecticasestenteters? 71 Monophyly ont belPhalaGromiynnxeGini tires. ....... seve edeevie-eunasachadvied.csddcenaatde<stmteatOeses taeversse 72
Apomorphies of tribe Phalacromyrmecini Potential synapomorphies between pairs of component tribes .............
Potential synapomorphies of Dacetonini + Phalacromyrmecini ...... er Ae Potential synapomorphies of Basicerotini + Phalacromyrmecinl ........2:cc0cccseeeeeesereeseeseerees 73 Potential synapomorphies of Dacetonini + Basicerotint .............:cceccesscssceeecssscsecsneeseeseeseneens 73 PRNGIXHSDECIESICISSE CTEM cap x rscucveer ceva, et owsdcvescsecstee ree otece ee wetanU NT Seas easee toe Saas 73 PNORMOWIEG SEMEN Grae chievncbecs ced ek Mk dex Lewap vas denscaxncsccazudssaueediens tian uea ta payed atctruces sasaverasisescaderiua tte 73 FRE LELEMICE SV Me uh hue att Rene ho PME 2G nnnesabhas Styne mnaeen CA ena dd Made utente ana eces ties easeuatnes 74
SYNOPSIS. Within the large subfamily Myrmicinae three independent tribes of ants, Basicerotini, Dacetonini and Phalacromyrmecini, are each recognised as monophyletic and are shown together to constitute a monophyletic tribe-group, here termed the dacetonine-group. Synapomorphies uniting the tribe-group and autapomorphies isolating each tribe are identified and discussed; comments on potential synapomorphies among the three tribes are given.
done little to clarify the overall position, although the latter does indicate a number of potentially interesting groupings. An investigation of myrmicine compara- tive morphology which is currently taking place is
INTRODUCTION
For many years it has been obvious that the classifica-
tion of the ant subfamily Myrmicinae at tribe rank is inadequate, with genera or groups of genera being associated on flimsy evidence or merely on superficial similarity. The most recent synoptic classifications (Holldobler & Wilson, 1990; Bolton, 1994) only serve to stress the rather decrepit nature of the structure. Detailed analysis of the subfamily to produce an accu- rate higher classification at this rank is in its infancy. Past work on myrmicine larvae (G. C. Wheeler & J. Wheeler, 1976) and sting structure (Kugler, 1979) has
© The Natural History Museum, 1998
unearthing many characters of value in establishing a natural tribe-rank classification. The aim of this paper is to establish, through morphological characters, the monophyly of three tribes within the Myrmicinae, namely Basicerotini, Dacetonini and Phalacromy- rmecini, and to show that the three together form a monophyletic group.
Most members of all three tribes nest and forage in leaf litter, topsoil or rotten wood where they form small, usually monogynous, colonies. Only few for-
66
age openly on the surface of the ground and few are arboreal or occur very deep in the soil. All known species are predatory, mainly catching entomobryomorph Collembola but with numerous spe- cies also preying on a wide range of other small arthropods such as sminthurid Collembola, Diplura, Symphyla, Chilopoda, Pseudoscorpiones, Acarina, Araneae, Isopoda, Amphipoda, and many orders of small Insecta and their larvae (Wilson, 1953, 1956; Carlin, 1982; Masuko, 1985; Wilson & Brown, 1985; Dejean, 1987a, 1987b and included references). Spe- cies and individuals are common in Berlese or Winkler bag samples of leaf litter or rotten wood throughout the tropics. They may be numerous or locally abundant (e.g. Belshaw & Bolton, 1994; Fisher, in press).
Dacetonini is a large tribe with a world wide distri- bution. At the last count it included 395 described species (Bolton, 1995a) but many more species have been diagnosed since and their taxonomy is currently being studied by the author. The vast majority of dacetonine species are tropical or subtropical but the tribe is also well represented in zones with a mediterranean climate. Some species occur endemically as far north as Switzerland and Georgia in the West Palaearctic (Kutter, 1977; Arakelian & Dlussky, 1991), and Japan in the East Palaearctic (Morisita, Kubota, et al., 1992). In North America one species ranges as far north as the USA — Canada border (Brown, 1953a). In the southern hemisphere dacetonine species occur as far south as New Zealand (Brown, 1953b), the southernmost parts of South Africa (Bol- ton, 1983), and central Argentina (Brown, 1962).
Basicerotini mostly shows a Gondwanic distribu- tion, with many species occurring in the Neotropical and Australasian regions (Brown & Kempf, 1960). In the New World basicerotines have been recorded as far north as Florida in the USA (Deyrup, Johnson, et ai., 1989) and as far south as Argentina (Kempf, 1972). The tribe is also well represented in Indonesia, Malay- sia, the Philippines and the Pacific islands (Taylor, 1990) but is absent from the Holarctic, Afrotropical, Malagasy and Oriental regions. Currently there are 64 described species (Bolton, 1995a), but more await description in various collections.
Phalacromyrmecini is a widely distributed but small and apparently relict tribe that contains only 3 de- scribed species (Bolton, 1984), each comprising a monotypic genus. Two of these have only been col- lected once. They have been found in Brazil, Madagascar and Malaysia.
The three tribes together thus have 462 described species, representing over 10% of the subfamily Myrmicinae, itself easily the largest subfamily in the family Formicidae. The Myrmicinae, which contains 4377 species, represents over half the total described for the entire family. The monophyly of Myrmicinae is probably certain and is documented elsewhere (Baroni
B. BOLTON
Urbani, Bolton & Ward, 1992). Bolton (1994) gives a modern definition of the subfamily and keys to its genera, and all its species are catalogued in Bolton (1995b).
TAXONOMIC HISTORY
Dacetonini is the oldest suprageneric name in the group, initiated by Forel (1892) to include the genera (in the order that he gave them) Strumigenys, Orectognathus, Epitritus, Hypopomyrmex [a fossil taxon], Daceton, Acanthognathus, Rhopalothrix, Ceratobasis [now a synonym of Basiceros] and Cataulacus. By modern reckoning this is a fairly disparate group and Forel gave no definitive diagnosis. The next year Forel (1893) listed these same genera, as Myrmicinae tribe Dacetonii, in a synoptic classifica- tion. Emery (1895) also produced a synopsis in which he rendered the tribe name as Dacetii. He did not mention the fossil Hypopomyrmex and rightly ex- cluded Cataulacus from the tribe. The first formal diagnosis of Dacetonini was produced in key form by Emery (1896) with the above inclusions and exclu- sions. The tribe remained stable with these seven genera for a number of years (W. M. Wheeler, 1910).
In later synopses and classifications other genera, described in the intervening years and referred to Dacetonini, were added, for instance Stegomyrmex, Microdaceton, Pentastruma by the time of Emery (1914), who spelled the tribe name as Dacetini; Glamyromyrmex and Epopostruma by the time of Forel (1917); Blepharidatta by the time of Emery (1922); Codiomyrmex by the time of W. M. Wheeler (1922), who also correctly excluded Stegomyrmex and Blepharidatta.
Over the twenty years that followed this a number of authors added further genera to the tribe, until by 1945 the number of genera stood at about 21. Shortly after- wards Brown (1948) began a series of revisionary studies on the generic composition of the tribe, adding a number of new genera. Subsequent papers (Brown, 1949a, 1949b, 1949c, 1950a, 1950b, 1952, 1953a, 1954) refined his concept of the tribe and its compo- nent genera. Several more new genera were described, some were synonymised, a number of incorrectly placed taxa were excluded, and the ranks of several genus-group taxa were adjusted. The tribe was subdi- vided into five subtribes (Brown, 1952) and a proposed evolutionary sequence within the tribe was produced (Brown & Wilson, 1959). A very old name, the fossil Hypopomyrmex that had been one of the original members of the tribe (and sole member of subtribe Hypopomyrmiciti), was finally reassesed (Brown & Carpenter, 1979) and decisively excluded.
In one paper of his revisionary series Brown (1949c) recognised that a number of closely related genera
MONOPHYLY OF THE DACETONINE TRIBE-GROUP
formed a uniform group that could be excluded from Dacetonini. These genera (Acanthidris, Basiceros, Creightonidris, Heptastruma, Octostruma, Rhopalothrix and Talaridris in the 1949 paper) were grouped together as the tribe Basicerotini. The species of all these were later fully revised on a world basis by Brown & Kempf (1960).
At about the same time Kempf (1960) described an oddly dacetonine-like genus (Phalacromyrmex) which together with two other small genera was diagnosed as a discrete genus group by Bolton (1984). A tribe rank name, Phalacromyrmecini, was applied to this group by Dlussky & Fedoseeva (1988), citing the Bolton (1984) definition. [Earlier uses of the name Phalacromyrmecini by G. C. Wheeler & J. Wheeler (1976, 1985) are unavailable as they do not satisfy Article 13 (a) of the International Code of Zoological Nomenclature, third edition, 1985.]
Thus by the early 1990s the classification and ge- neric content of the three tribes had reached the state outlined below, full references for which can be obtained from Bolton (1995b). The synopsis given here indicates all names in the genus group that are currently included in the three tribes. The names listed in parentheses are current junior synonyms. Figures in square brackets after the tribe names indicate the number of species examined for characters visible without dissection; a list of species dissected is ap- pended below.
Tribe Dacetonini [569]
Subtribe Dacetoniti [5] Genera: Acanthognathus, Daceton (=Dacetum).
Subtribe Epopostrumiti [15] Genera: Colobostruma (=Alistruma, =Clarkistr- uma), Epopostruma (=Hexadaceton), Mesostruma, Microdaceton.
Subtribe Orectognathiti [14] Genus: Orectognathus (=Arnoldidris).
Subtribe Strumigenyiti [535] Genera: Asketogenys, Chelystruma, Cladarogenys, Codiomyrmex, Codioxenus, Dorisidris, Dysedrognathus, Epitritus, Glamyromyrmex (=Borgmeierita), Gymnomyrmex, Kyidris (=Polyhomoa), Neostruma, Pentastruma, Quadristruma, Serrastruma, Smithistruma (=Cephaloxys, =Miccostruma, =Platystruma, =Weberistruma, =Wessonistruma), Strumigenys (=Eneria, =Labidogenys, =Proscopomyrmex, =Pyramica), Tingimyrmex, Trichoscapa.
Tribe Basicerotini [49]
Genera: Basiceros (=Ceratobasis, =Aspididris), Creightonidris, Eurhopalothrix, Octostruma, Protalaridris, Rhopalothrix (=Acanthidris, =Heptastruma), Talaridris.
67 Tribe Phalacromyrmecini [3]
Genera: Ishakidris, Phalacromyrmex, Pilotrochus.
Until 1994 characters defining the tribes tended to be of an inclusive nature. They were organised in such a way that all taxa showing a particular combination of characters were included in the tribe, those not show- ing such a combination were excluded. Usually none of these characters could be obviously pointed out as apomorphic. Along these lines the most recent inclu- sive definitions of the tribes were Brown (1953a) for Dacetonini, Brown & Kempf (1960) for Basicerotini and Bolton (1984) for Phalacromyrmecini.
Baroni Urbani & de Andrade (1994) identified the first synapomorphy exhibited by, and thus unifying, all three tribes: opposing, as distinct from overlapping, mandibles. They also demonstrated that some charac- ters regarded in the past as potential synapomorphies (head shape, specialised hairs) had no real value. Failing to find apomorphies for the individual tribes within the bounds of their synapomorphy, they termi- nated the independent existence of Basicerotini and Phalacromyrmecini, treating them as junior synonyms of Dacetonini. Although the Baroni Urbani & de Andrade (1994) position was summarily reversed in Bolton’s (1995b) catalogue, their approach provided a useful starting point for a detailed investigation. Based on the result of their work, the hypothesis postulated was that only a single real tribe is represented. Charac- ters were then sought and analysed to validate or disprove the results of that assumption. The results obtained indicate that the hypothesis is incorrect and that three tribes should be recognised. The diagnosis and discussion of the relevant characters follows this section.
The present investigation is based upon the worker caste, although most if not all of the characters men- tioned are also applicable to known queens. Males are excluded from the survey for the simple reason that they remain utterly unknown in phalacromyrmecines and in several genera of the other tribes. Even in the larger genera males are only very scantily represented. This means that selection of universal characters can- not even be implied, let alone be guaranteed. For the sake of this paper each currently recognised genus in the classification outlined above is assumed to be a valid taxon at genus rank, although work in progress indicates that many should not have such status.
MONOPHYLY OF THE DACETONINE TRIBE-GROUP
The tribes Basicerotini plus Dacetonini plus Phalacromyrmecini share five worker/queen synapo-
68
morphies and therefore together form a monophyletic group within the subfamily Myrmicinae. The synapo- morphies are listed and discussed below, each under its own heading. Character polarity here and in subse- quent sections is determined by comparison with those groups currently recognised as the most morphologi- cally generalised Myrmicinae (Myrmicini, Tetramori- ini, Formicoxenini). Characters and states across the entire subfamily have been taken into consideration to account for cases of convergence.
Synapomorphies of the dacetonine tribe- group
Masticatory margins of mandibles oppose but do not overlap at full closure Throughout the three tribes the masticatory (inner) margins of the mandibles fail to overlap or cross over when the mandibles are fully closed. In taxa with serially dentate mandibles the main dental rows inter- lock tightly at full closure (Figs. 5, 10, 15, 23) rather than the dental row on one mandible passing over the other (Figs. 1, 2, 4), although a few teeth at the extreme apices may cross over (Figs. 11, 22). In taxa with elongate or linear mandibles the inner margins oppose each other, or touch for part of their length at full closure, but do not overlap (Figs. 13, 14, 24). In such forms where a long preapical spiniform tooth occurs, or where an apical fork of spiniform teeth is devel- oped, the teeth themselves may interlock or cross, but the margins from which they arise do not do so (Figs. 13, 14, 33).
This is the character first recorded by Baroni Urbani & de Andrade (1994) and is apomorphic through the groups under consideration. With one exception other Myrmicinae, including all the tribes regarded as mor- phologically most generalised on other grounds (Myrmicini, Tetramoriini, Pheidolini, Formicoxenini), have the mandibles plesiomorphically overlapping at full closure (Figs. 1, 2, 4). The exception is the peculiar monotypic genus Tatuidris (Fig. 6), the sole extant member of tribe Agroecomyrmecini, which also has opposing mandibles. That this is a non-homologous parallelism with the dacetonine-group is easily dem- onstrated.
Tatuidris has the following apomorphies; the corre- sponding plesiomorphies exhibited throughout the dacetonine-group are given in square brackets.
1. Tergite and sternite of first gastral segment (ab- dominal segment 4) are fused. [First gastral tergite and sternite unfused. ]
2. First gastral sternite reduced, very much shorter than tergite, so that first gastral segment is directed ventrally with respect to the postpetiole. [First gastral sternite full-sized; first gastral segment not directed ventrally. ]
B. BOLTON
3. Mandible with a thick brush of stout setae on inner surface. [Mandible without setal brush. ]
4. Eyes at extreme posterior apex of scrobe. [Eyes not at posterior apex of scrobe.]
5. Antennal sockets extremely widely separated. [Antennal sockets relatively close together. ]
6. Alitrunk very short and compact, in profile higher than long. [Alitrunk elongate; in profile usually longer than high. ]
7. Apicotibial brush of thick dense setae present on foretibia opposite strigil. [Apicotibial brush ab- sent. ]
Conversely Tatuidris, beside lacking other dacetonine-group apomorphies, has the following char- acters exhibiting plesiomorphic states; their corresponding apomorphic expressions in the dacetonine-group (and mostly elsewhere in the Myrmicinae) are in square brackets.
1. Postpetiole very large, very broadly articulated to gaster. [Postpetiole small and narrowly articulated to gaster.]
2. Sternite of postpetiole large, in profile almost as extensive as tergite. [Sternite of postpetiole re- duced, much smaller than tergite. ]
3. Petiole sessile. [Petiole pedunculate. ]
4. Tibial spurs pectinate, strongly developed on mid- dle and hind legs. [Tibial spurs not pectinate, vestigial to absent on middle and hind legs. ]
Anterior head capsule is narrowed from side to side In the three dacetonine-group tribes the head in full- face view, anterior to the antennal sockets, is relatively narrow. A line drawn parallel to the long axis of the head, touching the outermost point of the torulus, will pass outside or very close to the point where the outer margin of the fully closed mandible intersects the anterior clypeal margin (Figs. 5, 7, 10, 13, 14, 15, 22, 23, 24, 33). In all other Myrmicinae (except for the few with radically migrated toruli discussed below) a line so constructed will pass considerably mesad of the point where the outer margin of the fully closed man- dible intersects the anterior clypeal margin (Fig. 1). The anterior narrowing of the head can be expressed as aratio, the Mandibular-Torular Index (MTI), which with the head in full-face view may be defined as: distance between points where outer margins of fully closed mandibles intersect anterior clypeal margin divided by distance between outermost points of lower margins of toruli. Thus when MTI =1.0 the two meas- urements are equal. MTI range of 0.5—1.3 is shown by the dacetonine-group of tribes, and also by Tatuidris (Fig. 6), Cataulacus (Fig. 4) and some Cephalotini. Other tribes of Myrmicinae together have MTI range 1.5—>3.0. The reason why these three non-dacetonine taxa have a similar MTI range to the dacetonine-group
MONOPHYLY OF THE DACETONINE TRIBE-GROUP
is that whilst their heads have remained plesiomorphically broad anteriorly, their toruli have secondarily migrated outwards on the cephalic dor- sum, bringing them into line with the clypeal intersection points of the mandibles. Thus two com- pletely different evolutionary routes, static toruli combined with narrowed anterior head (the dacetonine- group), versus static broad anterior head combined with outwardly migrated toruli (Tatuidris, Cataulacus, some Cephalotini), will give the same overall MTI result. For an illustration of this character spanning the entire subfamily compare the full-face view photo- graphs of the various myrmicine genera in Bolton (1994).
As a direct result of the narrowing of the anterior head the dacetonine-group shows a number of other derived characters that are unavoidable corollaries of this modification.
1. Width of labrum at base, and consequently also of clypeo-labral hinge, is reduced (see below).
2. Width of buccal cavity, and consequently also of labio-maxillary complex, is reduced (compare Fig. 3 with Figs. 12, 26, 28).
3. Anterolateral surface of head tends to be vertical, or nearly so, and is very close to or more or less directly below the antennal socket (e.g. Figs. 11, 182 19532):
Preocular carina is present With the head in profile or dorsolateral view there is a ridge, carina or lamella that originates at the posterola- teral termination of the clypeus, extends posteriorly below the antennal socket and usually continues for some distance towards the level of the eye (Figs. 11, 18, 19, 30, 32). In many taxa this preocular carina is also visible in full-face view, at least anteriorly (Figs. 13, 23, 25, 31), but usually cannot be seen in this view in those species or genera where the frontal lobes and frontal carinae are strongly expanded laterally (Fig. 7). Morphologically the carina represents the exagger- ated outer margin of the antennal fossa and is another corollary of the narrowing of the head (particularly of number 3, immediately above). At its weakest the carina 1s a ridge that emphasises the dorsalmost line of the more or less vertical side of the head capsule. At its strongest it is a broad, laterally projecting lamella. The structure is perhaps better termed the subtorular carina or subantennal carina as this area is always where it is strongest developed, but earlier literature has used preocular carina so much that the name will probably be retained.
Specialised range-finder/trigger hairs are developed on the mouthparts
In the dacetonine-group hunting always seems to in- volve an open-mandible approach to the prey, followed
69
by a rapid strike with the mandibles (see references above, particularly Masuko, 1985). The strike may be followed by static pressure of the mandibles, merely to retain a grip on the prey until the sting can be brought into use (‘strike-hold-sting’ technique), or dissipation of the kinetic energy of the strike itself may be suffi- cient to shock the prey into immobility, even if only temporary, until the sting can be brought into play, particularly if the prey is lifted clear of the ground immediately after the strike (‘strike-lift-sting’ tech- nique). Whatever the technique, the explosive closing of the mandibles is initiated by activation of a trigger which consists of one or more specialised hairs on the mouthparts. Contact with prey by these hairs com- mences a sequence of events that causes the mandibles to snap shut (Masuko, 1985; Gronenberg, 1996). In the vast majority of dacetonine-group taxa these hairs arise from the anterior margin of the labrum or the apices of the labral lobes (Figs. 16, 25-31 (broken off short in 26), 27, 33), more rarely from the mandible itself; such trigger hairs are not found elsewhere in the Myrmicinae.
Specialisation of the labrum
Except for the dacetonine-group of tribes the structure of the labrum is very uniform and generalised in the Myrmicinae (Gotwald, 1969 and present investiga- tion), and this same structure is prevelent throughout the Formicidae. Outside the dacetonine-group the la- brum is a simple sclerite (Fig. 3) that is broader than long and broadly hinged to the clypeus. Its posterior margin, which is attached to the clypeus, is more or less straight. Its anterior (free) margin is indented or cleft medially so that in general the labrum appears broadly B-shaped, or D-shaped with a median inden- tation in the anterior free margin. The clypeo-labral hinge is mobile so that at rest the labrum folds back and down with respect to the clypeus. In this position it tightly overlaps and protects the delicate apical portions of the labio-maxillary complex. In the Myrmicinae this is the plesiomorphic shape and posi- tion. It is encountered in all the morphologically more generalised tribes (e.g. Myrmicini, Tetramoriini, Pheidolini, Pheidologetonini, Formicoxenini, Solenopsidini), as well as in those that are rather more specialised morphologically (e.g. Attini, Cataulacini, Cephalotini, Crematogastrini, Stenammini, Agroecomyrmecini).
Tribes in the dacetonine-group show a number of modifications away from the generalised structure illustrated in Fig. 3, but all are based on labral narrow- ing and elongation, and loss of the basic broadly B- or D-shaped outline.
In taxa with mandibles that use static pressure, regardless of whether the mandibles are long or short, the labrum becomes longer and narrower, sometimes longer than broad. The sclerite usually terminates in
70
one or two exaggerated anterior lobes or linguiform prominences that bear the trigger hairs (Figs. 12, 16, 25, 28, 29, 31). In these forms the labrum can usually still flex down very slightly, but it is not capable of tight closure against the labio-maxillary complex (Figs. 12, 28). Conversely, in some taxa the labrum is hyper- trophied and elongate-linguiform; instead of concealing only the apex of the labio-maxillary complex it covers the entire buccal cavity (Fig. 32). Taxa with long kinetic mandibles tend to have the labrum very narrow basally, and more distally to develop a pair of laterally projecting processes or arms; the sclerite therefore tends to be roughly T-shaped and its apex functions to prop open the mandibles prior to striking (Figs. 26, 27, 30). One long-mandibulate genus (Acanthognathus) has eliminated the labrum from this function and here it is represented only by a very slender, inverted Y- shaped sclerite.
Some characters considered but not used at tribe-group rank
Presence of 2-segmented antennal club
An antennal club of two segments is universal in the tribes Basicerotini, Dacetonini and Phala- cromyrmecini, though only very poorly expressed in Daceton. This may be the result of a single evolution- ary event in the dacetonine-group, but a two-segmented club is also developed in a wide range of non- dacetonine myrmicine taxa. For example, a strongly defined two-segmented club is universal in Melissotarsini (Melissotarsus, Rhopalomastix), present in all the core genera of Pheidologetonini (Afroxyidris, Carebara, Oligomyrmex, Paedalgus, Pheidologeton), present in some but by no means all Stenammini (Adelomyrmex, Baracidris, Lachnomyrmex, Mayriella, Tetheamyrma), occurs in a few Solenopsidini (Carebarella, Solenopsis) and is present in the single extant genus of Agroecomyrmecini (Zatuidris). It is weakly developed in some Blepharidattini (Blepharidatta, Wasmannia), and in some but not all species of Cardiocondyla (Formicoxenini) and Crematogaster (Crematogastrin1).
Antennomere count
Although very useful as acharacterin keys, antennomere count was ignored in this survey because in some myrmicines there is variation within genera (e.g. 10, 11 or 12 in Tetramorium (Tetramoriini); 4, 5 or 6 in Smithistruma (Dacetonini)). Also, within the tribes of the dacetonine-group there is a wide range of counts that makes analysis very difficult. For example in Dacetonini counts of 4, 5, 6 and 11 have been recorded; in Basicerotini 7,8,9 and 12, andin Phalacromyrmecini 8, 9 and 11. Iam convinced that these counts indicate independent morphoclinal reductions within each tribe, with no significance at tribe rank.
B. BOLTON
Absence of tibial spurs
Throughout the tribes of the dacetonine-group spurs are absent from the middle and hind tibiae. This condi- tion is apomorphic among the Myrmicinae but unfortunately is so extremely widespread through the subfamily that it has very little analytical value in the current survey.
MONOPHYLY OF TRIBE BASICEROTINI
Basicerotini Brown
Basicerotini Brown, 1949c: 86. Type-genus: Basiceros Schulz, 1906: 156. [Basicerotini relegated as junior synonym of Dacetonini by Baroni Urbani & de Andrade, 1994: 10; revived from synonymy by Bolton, 1995b: 9.]
Apomorphies of tribe Basicerotini
Each worker apomorphy is briefly described, in some cases with added comments where unrelated taxa show similar but non-homologous modifications. The plesiomorphic state of each character is given in square brackets; in some cases alternative apomorphic condi- tions are also noted for comparative purposes.
LABRUM
Distal of the hinge with the clypeus the labrum has a deeply incised transverse groove or trench across its entire dorsal width. The distal margin of this groove is defined by a sharp edge, ridge or crest (Fig. 16; also visible between the closed mandibles in Figs. 13, 14). {Labrum without sharply defined transverse groove distal of hinge. ]
In the basicerotines this groove, or at least its sharp distal margin, is usually located far enough anteriorly on the labrum as to be visible in ordinarily mounted specimens which have the mandibles ajar and the labrum slightly depressed.
TORULUS
The dorsal lobe of the torulus is hypertrophied and strongly curved downwards (Figs. 18, 19); its outer surface is nearly vertical so that it conceals part to most of the condylar bulb of the scape and the antennal socket itself. [Torulus a simple annulus or with a small lobe present dorsally that is horizontal and does not conceal the condylar bulb of the scape and the antennal socket. |
SCAPE NECK ARTICULATION
Because of the size and shape of the torulus the space in which the basal neck of the scape (the short narrow section between condylar bulb and scape shaft proper)
MONOPHYLY OF THE DACETONINE TRIBE-GROUP
can move is narrow and directed fore and aft (Figs. 18, 19). [Space in which basal neck of scape can move is not restricted to a narrow fore and aft motion. ]
BASE OF SCAPE
Scape shaft near base is bent downwards through a right-angle or near right-angle; the articulatory condyle at the extreme base projects forward from this through another right-angle that is not in the same plane as the first bend but rather is rotated through about 90 degrees (Figs. 8, 9). [Scape shaft in a straight line with basal condyle; or if scape angled downward near base then angle of basal condyle remains in the same plane as the shaft and is not rotated through 90 degrees. ]
ANTENNAL FOSSA
Antennal fossa separated from scrobe by at least a cuticular rim or crest; the depressed fossal area sur- rounded on all sides by raised or prominent cuticle (Figs. 18, 19). [Antennal fossa and scrobe confluent (when the latter is developed). ]
SCROBE
Scrobe always present and located below the eye (Figs. 18, 19), the latter usually situated on the extreme dorsolateral rim of the scrobe, more rarely towards underside of upper scrobe rim (very rarely eye absent). [Scrobe absent, or present but extending above the eye (eye sometimes absent). ]
Some other myrmicine taxa have a scrobe that extends below the eye, namely Cataulacus, some Cephalotini, and the Dacetonini related toEpopostruma (Epopostrumiti in the synoptic classification above). Cataulacus and the few cephalotines with this condi- tion can be dismissed as obvious convergence. Not only is the basic structure of the scrobe dissimilar but also the detailed structure of the head is very different. They lack, of course, the apomorphies of the dacetonine-group of tribes and exhibit their own series of apomorphies. The Epopostrumiti, on the other hand, belong in the dacetonine-group. Apart from possess- ing the apomorphies of Dacetonini and lacking those of Basicerotini, the scrobes in Epopostrumiti lack sharply defined posterior margins, usually lack vent- rolateral margins and, except in a very few species, fail to extend forward to the mandibular articulation (Fig. 32); all of these are developed in Basicerotini. Finally, a morphocline of species in the genera Colobostruma — Mesostruma — Epopostruma is present that exhibits a gradual increase in development and definition of the scrobe; no basicerotine could be inserted into the series, nor added to either end.
OCCIPITAL FORAMEN
The occipital foramen is set in a deep depression on the occipital surface of the head; cuticular margination is continuous around the depressed area (Fig. 17). [Oc- cipital foramen not set in a continuously marginate deep depression. ]
Fl
A very few Dacetonini have a transverse ventral rim of cuticle below the occipital foramen. This structure is very different from the basicerotine organisation.
HELCIUM
Helcium arises from the base of a broad, deeply con- cave depression or excavation in the anterior surface of the postpetiole (Fig. 20). [Helcium not set in a concave depression. |
GASTER
First gastral tergite and sternite each distinctly trans- versely marginate basally, immediately behind the postpetiole (Fig. 21). [First gastral tergite and sternite not marginate basally. ]
Members of the strumigenyite group of Dacetonini genera have a specialised transverse crest on the first gastral tergite, the limbus, which is an apomorphy of that group. The limbus is located prebasally and is inclined towards the base proper, where it overhangs the presclerites of the segment. Itis nota homologue of the basal margination developed in Basicerotini. No Dacetoninihavea basally marginate first gastral sternite.
SCULPTURE First gastral tergite and sternite with characteristic sculpture of dense, sharply incised, separated punc- tures (e.g. Fig. 21). [First gastral tergite and sternite unsculptured, or with different sculpture. |
This character may seem rather imprecise, but the form of sculpture is striking and not obviously re- peated elsewhere. A very few individual basicerotine species have secondarily reduced or effaced the sculp- ture. Elsewhere in the dacetonine-group, and in the Myrmicinae as a whole, are many species with sculp- tured gasters. Even when the sculpture in these is of a basically punctate form it tends to be reticulate-punc- tate or sparse, or on one sclerite but not the other. It is not the dense deeply-incised punctation exhibited by the Basicerotini on both sclerites.
MONOPHYLY OF TRIBE DACETONINI
Dacetonini Forel
Dacetonini Forel, 1892: 344. Type-genus: Daceton Perty, 1833: 136.
Apomorphies of tribe Dacetonini
The two worker apomorphies are briefly described and comments added. The plesiomorphic state of each character is given in square brackets.
MANDIBLE Mandible with a medially projecting cuticular process present on the inner margin close to the base
72
(basimandibular process); the process not merely a modified tooth (Figs. 25, 26, 27, 31, 33). [Mandible without a basimandibular process. ]
In some dacetonine taxa the basimandibular process is visible in ventral view in ordinarily mounted speci- mens, but in most the mandibles need to be opened quite widely or the labrum depressed. The process is variably shaped in different dacetonine taxa, taking the form of a lobe, a short or long spur, or a lamella (basal lamella). In all dacetonines except Acanthognathus the basimandibular process inserts between the clypeus (above) and the labrum (below) when the mandibles are closed and is part of the jaw locking mechanism. In Acanthognathus the process is hypertrophied but passes ventral to the labrum as the latter is vestial in this genus and no longer serves in the jaw locking mechanism.
A few basicerotine species in the genera Eurhopalothrix and Octostruma have the basal tooth of the mandible flattened or lengthened. This is not homologous with the dacetonine basimandibular proc- ess, which is derived from the mandible itself and not from a tooth. Basicerotine species with a modified basal tooth are exceptional and certainly best regarded as independent acquisitions. The usual condition is to have the basal tooth quite normal; presence/absence of a flattened basal tooth cannot be used diagnostically either at genus or species-group rank.
LABRUM Dorsal surface of labrum with an impression or pair of impressions located medially on the labral shield, distal of the basal hinge but proximal of the labral lobes (Figs. 27, 30). [Labrum without mid-dorsal im- pression. |
The labral impression receives the basimandibular processes of the mandibles when they are fully closed. The character is not repeated anywhere else in the Myrmicinae. It is secondarily lost in Acanthognathus because, as pointed out above, the labrum is secondar- ily extremely reduced in this genus.
MONOPHYLY OF TRIBE PHALACROMYRMECINI
Phalacromyrmecini Dlussky & Fedoseeva
Phalacromyrmecini Dlussky & Fedoseeva, 1988: 80 [based on diagnosis in Bolton, 1984: 381]. Type- genus: Phalacromyrmex Kempf, 1960: 89. [Phalacromyrmecini relegated as junior synonym of Dacetonini by Baroni Urbani & de Andrade, 1994: 10; revived from synonymy by Bolton, 1995b: 9.)
B. BOLTON
Apomorphies of tribe Phalacromyrmecini
Each worker apomorphy is briefly described and com- ments added. The plesiomorphic state of each character is given in square brackets.
MANDIBLE
Dentition of alternating large and small teeth from base to apex; largest tooth usually the basal. [Dentition not of alternating large and small teeth; largest tooth usually the apical. ]
A few isolated species or small species-groups in the dacetonine genus Glamyromyrmex have the basal tooth the largest on the masticatory margin, but these lack alternating dentition.
MESOPLEURON
Katepisternum with an impression or groove extend- ing obliquely downward from posterior margin of mesopleural hair-filled gland towards metapleuron; impressed area usually bounded by ridges or carinae. [Katepisternal oblique groove absent. ]
SCAPE Scape slender basally, clavate apically; entire scape roughly Indian-club shaped. [Scape subcylindrical.]
POTENTIAL SYNAPOMORPHIES BETWEEN PAIRS OF COMPONENT TRIBES
The object of this paper has been to establish the monophyly of the tribes and the tribe-group, rather than to produce a formal phylogeny. However, anumber synapomorphies potentially linking pairs of tribes within the group have been noticed and these are mentioned below. The plesiomorphic state expressed by the isolated tribe in each set is given in square brackets.
Potential synapomorphies of Dacetonini + Phalacromyrmecini
POSTPETIOLE-GASTER ARTICULATION
Diameter of presclerites of abdominal segment 4 (=first gastral segment) small and constricted so that the postpetiole-gaster articulation is relatively narrow. [Basicerotini: diameter of these presclerites broad so that the postpetiole-gaster articulation is relatively wide. ]
PRESCLERITE OF FOURTH ABDOMINAL SEGMENT Pretergite of abdominal segment 4 (=first gastral seg- ment) subtended by a short narrow neck-like constriction. [Basicerotini: pretergite subsessile to ses- sile.]
MONOPHYLY OF THE DACETONINE TRIBE-GROUP
BASIMANDIBULAR GLAND Basimandibular gland present. [Basicerotini: basimandibular gland absent. ]
This gland is variously developed in many genera of these two tribes (/shakidris, Microdaceton, Strumigenys, Glamyromyrmex); it is universally ab- sent from basicerotines. Its absence from some dacetonines (Orectognathus, Acanthognathus) may be secondary.
MESOPLEURAL GLAND Anterolateral angle of mesopleuron bears a hair-filled glandular structure set in an emargination of the rim of the sclerite. [Basicerotini: mesopleural gland absent. ] This supposed gland varies from absent to mas- sively hypertrophied in Dacetonini and Phalacromyrmecini. It is universally absent in Basicerotini.
Potential synapomorphies of Basicerotini + Phalacromyrmecini
BASE OF MANDIBLE
Base of mandible with a long stiff ventrally directed seta on ventral margin. [Dacetonini: such a seta ab- sent. ]
This specialised seta is present in all Basicerotini. In Phalacromyrmecini it is obvious in /shakidris. Not recorded in the single specimen of Phalacromyrmex currently available, but this is in poor condition and badly mounted.
PROPODEAL SPIRACLE Propodeal spiracle low on side of sclerite, abutting the margin of the small metapleural gland bulla. {Dacetonini: propodeal spiracle high on side, widely separated from metapleural gland. |
In the dacetonine genus Acanthognathus the spiracle abuts the metapleural gland bulla but here the spiracle is high on the side and the bulla is secondarily ex- tended upwards.
Potential synapomorphies of Dacetonini + Basicerotini
None detected.
APPENDIX: Species dissected
The parts of all dissected specimens, whether partially or completely disarticulated, have been remounted on green-flagged card points and are deposited in The Natural History Museum, London. [List does not in- clude the many taxa mounted with mandibles open and mouthparts displayed, but not otherwise dissected; these have been mounted on blue-flagged card points in the Natural History Museum, London, collection. ]
Basicerotini
Basiceros. discigera, manni, militaris, singularis.
Eurhopalothrix: australis, biroi, bolaui, dubia, floridana, gravis, heliscata, insidiatrix, jennya, omnivaga, procera, punctata, speciosa, spectabilis, szentivanyl.
Octostruma: balzani, betschi, iheringi, inca, rugifera, stenognatha.
Protalaridris: armata.
Rhopalothrix: ciliata, isthmica, plus | unidentified species.
Dacetonini
Acanthognathus: brevicornis, ocellatus, rudis.
Codiomyrmex: thaxteri.
Colobostruma: alinodis, cerornata, plus | unidenti- fied species.
Daceton: armigerum.
Epitritus: argiolus, hexamerus, laticeps, roomi.
Epopostruma: frosti.
Glamyromyrmex: beebei, excisa, flagellatus, semicomptus, sistrurus, tukultus, plus 2 unidenti- fied species.
Gymnomyrmex: villiersi.
Kyidris: mutica, plus | unidentified species.
Mesostruma: browni, turneri.
Microdaceton: exornatum, tibialis, plus | unidentified species.
Neostruma: brevicornis, crassicornis, myllorhapha, zeteki.
Orectognathus: antennatus, clarki, mjobergi, szentivanyi, versicolor.
Pentastruma: sauteri.
Quadristruma: emmae.
Serrastruma: lujae, ludovici, serrula.
Smithistruma: alberti, angulata, dohertyi, fridericimuelleri, microthrix, ornata, transversa, truncatidens, plus 6 unidentified species.
Strumigenys: biolleyi, chyzeri, denticulata, doriae, elongata, gundlachi, godmani, koningsbergeri, louisianae, micretes, nidifex, pallestes, perplexa, precava, prospiciens, rogeri, rukha, saliens, signeae, subedentata, tigris, trieces, trudifera, plus 12 unidentified species.
Trichoscapa: membranifera.
Phalacromyrmecini
Phalacromyrmex: fugax. Ishakidris: ascitaspis.
ACKNOWLEDGEMENT
My sincere thanks to Suzanne Lewis (Department of Entomology, The Natural History Museum) for taking the SEM photographs used to illustrate this paper.
74
REFERENCES
Arakelian, G. R. & Dlussky, G. M. 1991. Murav’i triby Dacetini SSSR. Zoologichesky Zhurnal 70 (2): 149-152.
Baroni Urbani, C. & de Andrade, M. L. 1994. First description of fossil Dacetini ants with a critical analysis of the current classifica- tion of the tribe. (Amber Collection Stuttgart: Hymenoptera, Formicidae. VI: Dacetini.) Stuttgarter Beitrdge zur Naturkunde Serie B (Geologie und Palaontologie) 198: 1-65.
Baroni Urbani, C,. Bolton, B. & Ward, P. S. 1992. The internal phylogeny of ants. Systematic Entomology 17: 301-392.
Belshaw, R. & Bolton, B. 1994. A survey of the leaf litter ant fauna in Ghana, West Africa. Journal of Hymenoptera Research 3: 5-16.
Bolton, B. 1983. The Afrotropical dacetine ants. Bulletin of the British Museum (Natural History) (Entomology) 46: 267-416.
Bolton, B. 1984. Diagnosis and relationships of the myrmicine ant genus /shakidris, gen. n. Sytematic Entomology 9: 373-382.
Bolton, B. 1994. Identification Guide to the Ant Genera of the World: 222 pp. Harvard University Press, Cambridge, Mass.
Bolton, B. 1995a. A taxonomic and zoogeographical census of the extant ant taxa. Journal of Natural History 29: 1037-1056.
Bolton, B. 1995b. A New General Catalogue of the Ants of the World: 504 pp. Harvard University Press, Cambridge, Mass.
Brown, W. L. Jr. 1948. A preliminary generic revision of the higher Dacetini. Transactions of the American Entomological Society 74: 101-129.
Brown, W. L. Jr. 1949a. Revision of the ant tribe Dacetini. 3. Epitritus Emery and Quadristruma new genus. Transactions of the American Entomological Society 75: 43-51.
Brown, W. L. Jr. 1949b. Revision of the ant tribe Dacetini. 1. Fauna of Japan, China and Taiwan. Mushi 20: 1-25.
Brown, W. L. Jr. 1949c. Revision of the ant tribe Dacetini. 4. Some genera properly excluded from the Dacetini, with the establish- ment of the Basicerotini, new tribe. Transactions of the American Entomological Society 75: 83-96.
Brown, W. L. Jr. 1950a. Revision of the ant tribe Dacetini. 2. Glamyromyrmex Wheeler and closely related small genera. Trans- actions of the American Entomological Society 76: 27-36.
Brown, W. L. Jr. 1950b. Revision of the ant tribe Dacetini. 5. The delimitation of Arnoldidris new genus. Transactions of the Ameri- can Entomological Society 76: 143-145.
Brown, W. L. Jr. 1952. The dacetine ant genus Mesostruma. Transac- tions of the Royal Society of South Australia 75: 9-13.
Brown, W. L. Jr. 1953a. Revisionary studies in the ant tribe Dacetini. American Midland Naturalist 50: 1-137.
Brown, W. L. Jr. 1953b. A revision of the dacetine ant genus Orectognathus. Memoirs of the Queensland Museum 13: 84-104.
Brown, W. L. Jr. 1954. A preliminary report on dacetine ant studies in Australia. Annals of the Entomological Society of America 46 (1953): 465-471.
Brown, W. L. Jr. 1962. The Neotropical species of the ant genus Strumigenys Fr. Smith: synopsis and key to species. Psyche 69: 238-267.
Brown, W. L. Jr. & Carpenter, F. M. 1979. A restudy of two ants from the Sicilian Amber. Psyche 85 (1978): 417-423.
Brown, W. L. Jr. & Kempf, W. W. 1960. A world revision of the ant tribe Basicerotini. Studia Entomologica 3: 161-259.
Brown, W. L. Jr. & Wilson, E. O. 1959. The evolution of the dacetine ants. Quarterly Review of Biology 34: 278-294.
Carlin, N. F. 1982. Polymorphism and division of labor in the dacetine ant Orectognathus versicolor. Psyche 88 (1981): 231-244.
Dejean, A. 1987a. Behavioral plasticity of hunting workers of Serrastruma serrula presented with different arthropods. Sociobiology 13: 191-208.
Dejean, A. 1987b. Etude du comportement de prédation dans le genre Strumigenys. Insectes Sociaux 33 (1986): 388-405.
Deyrup, M., Johnson, C., Wheeler, G. C. & Wheeler, J. 1989. A preliminary list of the ants of Florida. Florida Entomologist 72: 91-103.
B. BOLTON
Dlussky, G. M. & Fedoseeva, E. B. 1988. Proiskhozhdenie i rannie etapy evolyutsii murav’ev (pp. 70-144). In Ponomarenko, A. G. Melovoi Biotsenoticheskii Krizis i Evolyutsiya Nasekomykh: 232 pp. Moskva: Nauka.
Emery, C. 1895. Die Gattung Dorylus Fab. und die systematische Eintheilung der Formiciden. Zoologische Jahrbiicher. Abtheilung fiir Systematik, Geographie und Biologie der Thiere 8: 685-778.
Emery, C. 1896. Clef analytique des genres de la famille des formicides. Annales de la Société Entomologique de Belgique 40: 172-189.
Emery, C. 1914. Intorno alla classificazione dei Myrmicinae. Rendiconto delle Sessioni della R. Accademia delle Scienze dell’'Istituto di Bologna (n.s.) 18: 29-42.
Emery, C. 1922. In Wytsman, P. Genera Insectorum. Hymenoptera, Fam. Formicidae, subfam. Myrmicinae. fase. 174: 207-397. Bruxelles.
Fisher, B. L. (in press). The Malagasy case. In Agosti, D. & Majer, J. (eds.). Measuring and Monitoring Biological Diversity: Standard Methods for Ground-living Ants. Smithsonian Institution Press, Washington, D.C.
Forel, A. 1892. Attini und Cryptocerini. Zwei neue Apterostigma- Arten. Mitteilungen der Schweizerischen Entomologischen Gesellschaft 8: 344-349.
Forel, A. 1893. Sur la classification de la famille des formicides, avec remarques synonymiques. Annales de la Société Entomologique de Belgique 37: 161-167.
Forel, A. 1917. Cadre synoptique actuel de la faune universelle des fourmis. Bulletin de la Société Vaudoise des Sciences Naturelles 51: 229-253.
Gotwald, W. H. Jr. 1969. Comparative morphological studies of the ants, with particular reference to the mouthparts. Memoirs of Cornell University Agricultural Experiment Station 408: 1-150.
Gronenberg, W. 1996. The trap-jaw mechanism in the dacetine ants Daceton armigerum and Strumigenys sp. Journal of Experimental Biology 199: 2021-2033.
Holldobler, B. & Wilson, E. O. 1990. The Ants: 732 pp. Harvard University Press, Cambridge, Mass.
Kempf, W. W. 1960. Phalacromyrmex, a new ant genus from south- ern Brazil. Revista Brasileira de Biologia 20: 89-92.
Kempf, W. W. 1972. Catalogo abreviado das formigas da Regiao Neotropical. Studia Entomologica 15: 3-344.
Kugler, C. 1979. Evolution of the sting apparatus in the myrmicine ants. Evolution 33: 117-130.
Kutter, H. 1977. Insecta Helvetica Fauna. 6. Hymenoptera, Formicidae: 298 pp. Fotorotar AG, Ziirich.
Masuko, K. 1985. Studies on the predatory biology of Oriental dacetine ants. 1. Some Japanese species of Strumigenys, Pentastruma, and Epitritus, and a Malaysian Labidogenys, with special reference to hunting tactics in short-mandibulate forms. Insectes Sociaux 31 (1984): 429-451.
Morisita, M., Kubota, M., Onoyama, K., Ogata, K., Terayama, M., Yamauchi, K., Sonobe, R., Yamane, S., Kondoh, M. & Imai, H.T. 1992. A guide for the identification of Japanese ants. 3. Myrmicinae and supplement to Leptanillinae. Myrmecological Society of Japan: 94 pp.
Perty, M. 1833. Delectus animalium articulatorum Fase. 3: 125— 224. Monachii.
Taylor, R. W. 1990. New Asian ants of the tribe Basicerotini, with an on-line computer interactive key to the twenty-six known Indo- Australian species. Invertebrate Taxonomy 4: 397-425.
Wheeler, G. C. & Wheeler, J. 1976. Ant larvae: review and synthesis. Memoirs of the Entomological Society of Washington 7: 1-108. Wheeler, G. C. & Wheeler, J. 1985. A simplified conspectus of the Formicidae. Transactions of the American Entomological Society
111; 255-264.
Wheeler, W. M. 1910.Ants: their structure, development and behavior: 663 pp. New York, Columbia University Press.
Wheeler, W. M. 1922. The ants of the Belgian Congo. Bulletin of the American Museum of Natural History 45: 1-1139.
Wilson, E. O. 1953. The ecology of some North American dacetine
MONOPHYLY OF THE DACETONINE TRIBE-GROUP 75
ants. Annals of the Entomological Society of America 46: 479-495. Wilson, E. O. & Brown, W. L. Jr. 1985. Behavior of the cryptobiotic Wilson, E. O. 1956. Feeding behavior in the ant Rhopalothrix biroi predaceous ant Eurhopalothrix heliscata, n. sp. Insectes Sociaux Szabo. Psyche 63: 21-23. 31: 408-428.
76 B. BOLTON
Figs. 12-22. Worker ants: 12, ventral head to show buccal cavity of Eurhopalothrix platisquama; 13-15, heads in full-face view of 13, Rhopalothrix ciliata, 14, Protalaridris armata; 15, Basiceros discigera; 16, oblique frontal view of head of Basiceros singularis, mouthparts extended and right mandible removed; 17, occipital foramen of Eurhopalothrix heliscata;
18-19, lateral view of head with antenna removed, of 18, Basiceros singularis; 19, Eurhopalothrix platisquama; 20,
frontal view of helcium of Eurhopalothrix procera; 21, base of gaster in oblique ventral view of Basiceros singularis; 22,
| head in full-face view of Dysedrognathus sp. n.
B. BOLTON
Figs. 23-33. Worker ants. 23-25, head in full-face view of 23, Smithistruma sp. n.; 24, Daceton armigerum; 25, Smithistruma reliquia, mandibles open; 26-28, ventral view of head to show buccal cavity of 26, Microdaceton sp. n., mandibles open; 27, Daceton armigerum, right mandible and labio-maxillary complex removed; 28, Smithistruma truncatidens;, 29-30, oblique frontal view of head to show labrum, mandibles removed, of 29, Smithistruma truncatidens; 30, Strumigenys sp. n.; 31, head in full-face view of Smithistruma kerasma, mandibles open; 32, head in ventrolateral view of Colobostruma sp.; 33, mandibles in oblique frontal view of Microdaceton tibialis.
Bull. nat. Hist. Mus. Lond. (Ent.) 67(1):79-152 Issued 25 June 1998
An annotated checklist of bumble bees with an analysis of patterns of description (Hymenoptera: Apidae, Bombini)
PAUL H. WILLIAMS
Department of Entomology, The Natural History Museum, Cromwell Road, South Kensington, London SW7 5BD
CONTENTS
| Hii OTG IC COLAC 01 SA SPR ee ORE nc RRRPEY cece ce ert pe SScR PPA Rae ORME 08 dc. ORE Sm ok Ore onan eee 80 PAS EALISE SEO IES IECLES socass vaaeteraccee dr caamevusa scaversussaeroncetaxcnes cade caps sauseizeccsee ured Tatars meen gone otsers = ane 80 FUISLOUY ON CIS COVEUY OL SPECIOS fences onseteteseascsncoseesasgatusch cust daceacoasy shes eatovedadec¥oates diospsoeesucece teas 80 ANSTO yal Chin UNIT GALCL OM OL TALES) « cecps deen ce <oeesk fab nes ch nectar asae cuca org <tsaccrab sad esexansuecheainssdcnancpesséee 81 Summary of historical and regional trends in describing bumble bees ...... 86 WEVElonmmemt Ola eVvISed CHECKING Opietececcrecct acecassevaraceseesasvedscceatitaactaroarccckacesseutnestacancecnencee 87 PMCMSIICHMLE CE CHIC IBS) nee tencerscerserrtcsestreteseraressteeesca ta esersgsefuxiurcrasteveseiasvadisesczsssesaosiaueeseaseescanes 88
SEONOMONY Meeteee tener cere ctu c cee cc M Scere ts settee acci ce sancancednevaessesved eaddzncs sassppocuyesneesagiai«spcd/seapenwascen 88 Phylogeny, supra-specific taxa and ordering Of SPeCieS ...........:eescceseesseeseeeeeeeeeeeeeneeeeeeeeeneees 89 Te MaNiGNALSOMMUMINALE SPECLOS ter Meech ics ccasconcectcascsescaccusscacussssussveccoassaseestvacensnetsatanesasnons 90 Sub-specific taxa
INomenGlatune?.2. ic .ccteit.cccees.s ATSIC) cred py kiteelll COMVEMIMOMS 4. meseee sa teeters csessxencsvevacespescaaecteatssdcscsesdeseccecddscesussvenrsdheustscceneers SEI SG HIT OV, 7900) PBB Ree eee eR oe se a ee A ee Rh nent et anh 97 MPU GATOS GLE ZN x oe, ete eae ean. on ccuvsschteederce acecceeracedees hl usctee MMe ctebenssecstlaedeaascs 97
IC YiS (ar oH iro) ol sTa FO) SA Reece a5 nt RR er AR ie ee Eo Fr ee ee 97
IBIS (ROIBSTIG CLES en eee eee, Soe Pe tearebeteeibe St... i LORE R aren Lt eden ee EI age oe EE 99
GRCKENGCES crrescess ve santh restr sd vansh cuban Mee ee Tyas oon saa acsht estate dade edassdeausseeshsasdddubendeealdsnseaeeteeeanea 140
SYNOPSIS. Bumble bees are among the minority of groups of organisms for which there is some evidence that most species have already been described. Nonetheless, a synoptic revision of the group has been delayed, in part by the difficulties imposed by an unusually high ratio of names to species (averaging more than | 1). To explore some of the factors contributing to this phenomenon, historical and geographical trends in the naming of bumble bees are summarised. This shows that most taxa were named by European authors, beginning with the most widespread European species, moving later to not only the more narrowly distributed species and to species from other parts of the world, but also to taxa at progressively lower nomenclatural ranks, particularly within the more widespread European species. Nearly half of all of these names have been published since the last world-wide checklist in 1922. In attempting to bring this up to date, the present checklist adopts broad interpretations of species and recognises a total of 239 recent species (including the social parasites but excluding fossil taxa), with 24 new synonyms and 29 provi- sional synonyms. The list also includes notes on alternative interpretations of taxonomic status and on nomenclatural problems, drawing attention to those cases where further research is most urgently needed. In particular, suggestions are presented for an application to the International Commision on Zoological Nomenclature to use its Plenary Power in order to conserve current usage of the commonly used names atratus, balteatus, distinguendus, flavifrons, humilis, hyperboreus, mesomelas, mixtus, norvegicus, polaris, pyrenaeus, soroeensis and variabilis.
© The Natural History Museum, 1998
80
INTRODUCTION
Bumble bees have long been popular with collectors. Just as with butterflies, part of the attraction may be explained by their bright colours, large body size, activity during daylight hours, and abundance in the north-temperate regions where most collectors have lived. As a result, large samples of bumble bee speci- mens have now been assembled, even from remote parts of the world.
A problem for biologists trying to identify bumble bee species, all the more apparent because of the large amount of material available, is that while bumble bees can be described as morphologically relatively ‘monotonous’ (Michener, 1990), they are often ex- traordinarily variable in the colour patterns of their pubescence. The situation is made worse by a strong tendency for species to converge locally on different colour patterns (Plowright & Owen, 1980).
Faced with this variation, generations of taxono- mists since the starting point of Linnaean nomenclature in 1758 have described differing individuals under a plethora of more than 2800 formal names (Williams, unpublished catalogue, including names for species, subspecies and synonyms, as well as infrasubspecific names, misspellings and other unavailable names). Most of these names are for taxa below the rank of species, and just 239 taxa are interpreted here as separate species. Arguably, the nomenclatural burden of more than 11 names per species (median 5, maxi- mum 186) has slowed progress towards a complete revision of the group. Hence there is a need for an overview which, although bound to require revision, will provide an improved framework for more detailed regional studies. It is also important to understand any regional or taxon-directed bias in patterns of taxo- nomic description when seeking to interpret patterns in diversity, ecology and biogeography. The present checklist begins to address these needs.
Past lists of species
There have been few attempts to present complete revisions, catalogues or checklists of all bumble bee species from which to see summaries of past views. Latreille (1809) included 13 species in his genus Bombus. Most of the early lists included just those species seen by their authors, usually from particular collections, and often from just one region. For exam- ple, Smith (1854) catalogued 87 bumble bee species (79 Bombus + 8 Apathus [= Psithyrus]) in the collec- tion of the British Museum. The only truly synoptic catalogue of bumble bees was published by Dalla Torre (1896), with 255 (non-fossil) species (228 Bombus + 27 Psithyrus). It included many varietal names, synonyms and early references. The reason
P.H. WILLIAMS
why Dalla Torre’s species count exceeds the total now recognised as described before 1899 (159 species, Fig. 1) is of course that many of his species are now treated as synonyms or subspecies. Later, Skorikov (1922a) listed 237 species (plus 70 ‘Bombi incertae sedis’ ), but with few synonyms and without including Psithyrus. Nonetheless, Skorikov’s list did arrange most of the known species within his genera and subgenera, which form the basis of the current subgeneric system (Richards, 1968).
Taken together, the few past lists of bumble bees show that the number of taxa accepted as species at a particular date grew rapidly during the nineteenth century, but has since remained relatively stable, with a slight decline to the conservative estimate of 239 species in the present list. Undoubtedly part of the explanation for this decline lies in the relatively con- servative species concept accepted at present (see below). This reflects a gradual shift in emphasis among criteria for recognising species from the use of colour characters to the use of morphological characters, particularly to using characters of the male genitalia (see the introductory comments by Radoszkowski, 1884). A similar pattern of growth and decline has been found for past numbers of milkweed butterfly species (Ackery & Vane-Wright, 1984). However, there might now be another period of rapid growth if mo- lecular characters and phylogenetic species concepts (discussed below) were to be applied (cf. discussion of the number of bird species by Martin, 1996; Patterson, 1996; Zink, 1996, 1997; Snow, 1997).
History of discovery of species
The dates of first formal description for the currently recognised bumble bee species show that the highest rates of species discovery were in the latter half of the nineteenth century through to the First World War (Fig. 1, median date 1877). These species are recog- nised retrospectively from the present list, rather than as the numbers accepted within each time period. The larger dips in overall rate of discovery may be associ- ated with factors such as war and its aftermath (e.g. Napoleonic and Second World Wars), presumably through constraints on resources and on freedom of travel.
Some authors described many more bumble bee species than others: 45% of presently accepted species were described by just 10% of the authors who de- scribed these species (Smith 32 species, Skorikov 19, Cresson 17, Morawitz 15, Radoszkowski 13, and Friese 12). Similarly, Gaston, Scoble & Crook (1995) found a skewed pattern of activity among authors describing geometrid moths. But of the six authors who described the most bumble bee species, only Ezra Cresson (Snr) actually worked in the New World, whereas the other five were based in Europe (including European Russia).
CHECKLIST OF BUMBLE BEES 250
81
200
150
Species
100
1750 1770 1790 1810 1830 1850
1870" 1890" 1810) ~ 1930" 1950" 1970 = 1990
Date of description
Fig. 1 Rate (lower grey) and cumulative number (upper white) of first formal descriptions of presently recognised bumble bee species (dates from the oldest available names in the sense of ICZN, 1985).
Rates of discovery of bumble bee species vary to some extent among biogeographic regions (Fig. 2). Again, this was also found for geometrid moths by Gaston, Scoble & Crook (1995). For bumble bees, the recent proportional discovery rates have been lowest in the New World and highest in the Oriental Region. The Neotropical Region appears to have a small known bumble bee fauna for its large area. There is no obvious evidence that descriptive effort has been lacking, al- though detailed revisionary work is needed and species with small range sizes may remain to be discovered. In contrast, the Oriental Region’s high recent propor- tional rate of species discovery, despite its smaller area than the other regions, is possibly explained in part because it has been studied intensively for a shorter period.
The world-wide rate of discovery of genuinely un- known bumble bee species appears now to be slowing down (Fig. 1). Undiscovered species are very likely to remain, although there is no evidence that large num- bers of species are awaiting description in collections (although some known subspecific taxa might yet be recognised as species if changes were to occur in species concepts or in the availability of character evidence, see Martin, 1996; Patterson, 1996; Zink, 1996). The sigmoidal pattern of species discovery in Fig. 1 isalsoshownby afew other relatively well known groups such as birds, although for most large groups (including Hymenoptera as a whole) the rates of de- scription continue to be high or are even increasing (Hammond in Groombridge, 1992; Tennessen, 1997).
History of publication of names
The present checklist is intended only to address the question of taxa at the rank of species (see below). For this purpose it is not necessary to consider concepts of taxa at the rank of subspecies and below and subspecific names may be treated in analyses as further synonyms of species (Gaston & Mound, 1993). This is not to say that subspecific taxa ought not to be recognised if they were considered useful in the context of other studies. In addition, some authors have applied classical names to taxa at even lower nomenclatural ranks, for example in referring to ‘varieties’ or ‘forms’ within subspecies. These are now interpreted as infrasubspecific names and are ‘unavailable’ for use in the sense of the /nter- national Code of Zoological Nomenclature (ICZN, 1985). They have had to be included in a manuscript catalogue (unpublished) in order to avoid confusion by explicitly resolving questions of nomenclatural status and availability. Infrasubspecific names are included in this analysis as a category separate from specific or subspecific names because of their particular signifi- cance for understanding historical patterns in the description of diversity at the lowest nomenclatural ranks.
Bumble bees have the highest known levels of synonymy (83%, or 92% if infrasubspecific names were to be included) in comparison with the range of insect taxa reviewed by Gaston & Mound (1993). Their results showed synonymy levels ranging from 7% for Siphonaptera to 80% for Papilionidae and
82
Palaearctic Region 120
80
40
0) 1750) 1790! 18380) 1870) 19110) 1950) 1'9S0
Oriental Region 60 40
20
0 1750) 177/90) 1880) 1870) 1910) 19505) 1390
P.H. WILLIAMS Nearctic Region
48 32
16
(0) 1750) 11790) «1830 1870" 1910) SaiS50)) S80
Neotropical Region
16
0 1750; > 1790), 1830) | 1870) 190 ISSO RT SS0
Fig. 2 Cumulative number of first formal descriptions of presently recognised bumble bee species with centres of area of occupancy (so species lists do not overlap) in each of the four principal biogeographic regions occupied by bumble bees (dates from the oldest available names in the sense of ICZN, 1985; regions defined in Williams, 1996b: fig. 1; Oriental includes northern and southern Oriental Regions; Nearctic includes northern, central and southern Nearctic Regions; Neotropical includes northern, central and southern Neotropical Regions; the Arctic Region is excluded; species that are exclusively peri-Tibetan Oriental but which nevertheless have range centres in Palaearctic central Tibet by simple range
averaging are included as Oriental species).
Pieridae combined. It must be born in mind that the insect taxa that they surveyed are all more speciose than the bumble bees by a factor of at least four, and extreme values for larger groups are less likely. None- theless, Gaston & Mound (1993) also noted that the two families of most brightly coloured butterflies have the highest levels of synonymy and that these families have many more subspecific names than the smaller and duller-coloured hesperiid butterflies. R. I. Vane- Wright (pers. com.) suggests that synonymy rates may be particularly high among the large, colourful butter- flies of the Danainae and Parnassius.
In contrast to the discovery of currently recognised species, the greatest activity in publishing names for all supposed bumble bee taxa at the rank of species and below was concentrated slightly later than for pres- ently recognised species, in the first half of the twentieth century (Fig. 3, median date 1922). This difference may be explained in part by the logical inevitability that synonyms and names for taxa below the rank of species can only be published subsequently to valid
species names (i.e. the oldest available names, exclud- ing junior homonyms, in the sense of ICZN, 1985). If these names were in effect to represent the redescription of known species at random, then the earlier described species might be expected to have accumulated more names. Studies of other taxa have also shown that both the date of first description and the number of syno- nyms per species may be affected by variation in the size of a species’ geographic range (as well as by other factors such as body size). Large range size is likely to affect the date of first description because it contrib- utes to a greater ‘apparency’ of the species to collectors (Gaston, Blackburn & Loder, 1995), particularly as broad correlations between range size and abundance suggest that widespread species also tend to have higher local densities (Brown, 1984; Gaston, 1994; for bumble bees, see Hanski, 1982; Williams, 1988).Apart from enhancing the chances of random redescription, large range size is also likely to affect the number of synonyms because there is a greater likelihood that specimens collected in one area will be regarded as
CHECKLIST OF BUMBLE BEES 3000
83
2500
2000
Names 3 +
1000 +
1770 1790 1810 1830 1850
1870 1890 1910 1930 1950 1970 1990
Date of description
Fig. 3 Rate (lower grey) and cumulative number (upper white) of all descriptions with classical names for bumble bee species, subspecies and infrasubspecies since the starting point of zoological nomenclature in 1758 (from a manuscript
catalogue, unpublished).
distinct from specimens collected from another distant area, because character variation is apt on average to be greater (Gaston, Blackburn & Loder, 1995).
For the bumble bee catalogue data, the number of synonyms (including subspecies, but excluding infrasubspecies) is correlated with both the date of first description and the range size of a species independ- ently of one another, although slightly more of the variation is explained by variation in range size (par- tial r, Table 1). Many of the species with large range sizes, early dates of first formal description and many synonyms are found in western Europe (i.e. triangles at the left and upper part of Fig. 4). Most of these species occur in either the lowland areas of Europe where early naturalists were most active, such as Britain, or else are nearly circumpolar in their distribu- tion.
Curiously, all of the infrasubspecific names (34% of all names as interpreted at present) belong to the bumble bee species of the Old World (Fig. 5). Species of the Old World also have more synonyms and sub- species per species than do the species of the New World (numbers of names log-transformed and ex- cluding 6 Holarctic species, 1,,,.= 3.81 with separate variance estimates, p< 0.001). | One possible explanation for the greater numbers of names per species for bumble bees of the Old World is that they might have broader distributions
than the species of the New World (see above). This could arise because the Old World has a slightly larger total area of suitable habitat (bumble bees oc- cupy 131 of the 611,000 km? grid cells in the Old World and 117 in the New), which is apparently subdivided into fewer well differentiated biogeographic assemblages of bumble bee species (e.g. Williams, 19965: fig. 1). However, this explana- tion is not strongly supported by the bumble bee data, which show the difference in range sizes between the two hemispheres to be not significant, (range sizes log-transformed and excluding 6 Holarctic species, t,,,= —1.24 with separate variance estimates, p= 0.22). Consequently, while an effect of differences in habitat area will deserve future consid- eration, other effects are likely to be more important.
A second possibility is that whereas bumble bee taxa of uncertain rank may have tended to be re- garded more often as subspecies in the Old World, in the New World they may have tended to be regarded as species (see the discussion below of criteria to recognise species). While this factor could have con- tributed to the observed patterns, it is unlikely to explain why (at a lower rank) so many infrasub- specific names were described exclusively for taxa from the Old World.
A third possibility is that the diversity of languages used for taxonomic publications in the Old World may
84
Table 1
P.H. WILLIAMS Results of multiple regression of numbers of synonyms/subspecific names (infrasubspecific names are excluded;
from a manuscript catalogue, unpublished) on date of first formal description for presently accepted species and range size
(number of occupied 611,000 km? grid cells world-wide). Partial r values indicate the correlations with the synonyms variable after adjusting for the other predictor variable in each case.
log, ,(synonyms+1) = 6.316(+0.969) — 0.003(+0.0005).date + 0.401(+0.057).log, (range) multiple = 0.72
Fg leo 16 p< 0.0001 partial r Ure Dp date —0.390 -6.51 < 0.0001 log,,(range) 0.418 7.08
< 0.0001
pascuorum a
lucorum s.]
4
guahuoukS
soroeensjs
&
terrestris
Fig. 4 Scatterplot of 239 presently accepted bumble bee species by range size (number of occupied 611,000 km? grid cells world-wide), date of first formal description and numbers of synonyms/subspecific names (infrasubspecific names are excluded; from a manuscript catalogue, unpublished). The British fauna is distinguished as filled triangles, the nearly
circumpolar fauna (B. hyperboreus, B. balteatus, B. polaris and B. lapponicus) as squares, and some British and widespread European species are labelled individually.
CHECKLIST OF BUMBLE BEES
85
1200 |
1000 |
800 }
600
Numbers
400
200
Old World
MMM Species Synonyms [= Infrasubspecies
New World
Fig.5 Number of presently recognised bumble bee species, synonyms/subspecific names and infrasubspecific names for the Old World and the New World (from a manuscript catalogue, unpublished).
have impeded communication and lead to more fre- quent re-description of taxa than in the New World, where English was much more dominant (C. O’ Toole, pers. com.). Again, while this factor is likely to have contributed to the observed patterns of synonyms, it does not explain why (at a lower rank) so many infrasubspecific names were described exclusively for taxa from the Old World.
Another possible interpretation, which might ex- plain more of the differences in description dates between Figs. 1-3 as well as the differences in the distribution of bumble bee subspecies, synonyms and infrasubspecific names between hemispheres (Fig. 5), is that during the twentieth century, effort for describ- ing the variety of these insects may have become, in effect, re-directed towards finer distinctions and lower nomenclatural ranks within known species. This is perhaps likely as undescribed species became inevita- bly more difficult to find close to home for the most active taxonomists, who were based in Europe. Three lines of evidence are consistent with this explanation. First, slightly more of the variation in richness of infrasubspecific names among species is accounted for by variation in the date of first description of the species (partial r, Table 2), rather than by variation in total range size. This is in contrast to the pattern for synonyms alone (cf. Table 1), although species that are sufficiently widespread in lowland Europe to include Britain within their distributions still tend to have high numbers of both synonyms and infrasubspecific names (Fig. 6, e.g. B. pascuorum, B. lucorum). A second intriguing observation is that compared to the number of authors who have published presently accepted species names, only one third the number of authors (20) have published infrasubspecific names, even though there are nearly four times as many infrasubspecific names. Indeed, just three of these authors (Bruno Pittioni, Edgar Kriiger and Alexander
Skorikov) are responsible for 70% of the infra- subspecific names (all of the species with many infrasubspecific names had been described before these three authors became active in publishing infra- subspecific names between 1910 and 1960, see Fig. 7). Many similar examples are known from work on butterflies (R. I. Vane-Wright, pers. com.), with au- thors choosing a particular favoured species and describing large numbers of infrasubspecific names (e.g. Bright & Leeds, 1938). The third point is that the three most prolific authors all worked in Europe, and there is a correlation among all 239 bumble bee spe- cies between the number of infrasubpecific names and the breadth of the species’ distributions just within Europe (measured as the number of occupied 611,000 km? grid cells between Britain and the Urals, but excluding Atlantic islands, North Africa, Turkey and the Caucasus; Spearman r= 0.67,1,,,= 13.99,p< 0.001). Thus, a high proportion of the many infrasubspecific names were published by very few European authors, for previously described species that are also particu- larly widespread in Europe.
High numbers of synonyms and infrasubspecific names for B. terrestris and B. lucorum (subgenus Bombus) and for B. humilis and B. pascuorum (subgenus Thoracobombus) in Fig. 6 raise the possi- bility that large numbers of names are associated with particular groups of species, perhaps with particular subgenera. Number of names per species is plotted against range size per species for subgenera in Fig. 8. These properties are correlated (log-transformed data, correlation r= 0.58, F, ,.= 18.16, p< 0.001), but it is the subgenera with high scores that are more informative. The subgenus Kallobombus includes many names, but only a single, very variable species B. soroeensis, which is broadly distributed in Europe (see below and Reinig, 1939: fig. 10). The subgenera Alpinobombus and Laesobombus also have broadly distributed
86
P.H. WILLIAMS
Table 2 Results of multiple regression of numbers of infrasubspecific names (from a manuscript catalogue, unpublished) on date of first formal description for presently accepted species and range size (number of occupied 611,000 km? grid cells world-wide). Species without infrasubspecific names were excluded from the analysis. Partial r values indicate the correlations with the infrasubspecific names variable after adjusting for the other predictor variable in each case.
log Ginfrasubspecifics+1) = 14.169(+3.638) — 0.007(+0.002).date + 0.742(+0.212).log, (range)
multiple r= 0.66
partial r date —0.364 log, (range) 0.340
eee 36:57 p< 0.0001
to4 P
-3.79 < 0.001 3.50 < 0.001
20 Ne —
140
120
100 7) ® 5 By 80 L Oo ® o Te) 60 >| @ = rupestris = ruderarius 40 s
pomorum
pratorum
20
0) 10 20 30
lucorum s.| a
terrestris A
soroeensis
a humilis A
pascuorum a
40 50 60 70 80
Synonyms
Fig.6 Scatterplot of 239 presently accepted bumble bee species by numbers of infrasubspecific names and numbers of synonyms/subspecific names (from a manuscript catalogue, unpublished). The British fauna is distinguished as filled triangles, the nearly circumpolar fauna (B. hyperboreus, B. balteatus, B. polaris and B. lapponicus) as squares, and some
widespread European species are labelled individually.
species but relatively few names, perhaps because they are absent or not abundant in those parts of Europe where the authors publishing most bumble bee names have worked, despite several of the species being very variable in colour pattern (e.g. B. balteatus). In con- trast, the high ratio of names per species for the subgenus Bombus shows the keen interest by some European authors such as Kriiger (1951, 1954, 1956, 1958) in describing the finer points of variation, not so
much within the North American species, but particu- larly within the widespread European species, B. terrestris and B. lucorum.
Summary of historical and regional trends in describing bumble bees
Based on the evidence of asymptotic tendencies in species-discovery curves, a higher proportion of all
CHECKLIST OF BUMBLE BEES 140
120
100 n oO § S 80 fc) i (Ss) oO a a 60 =I Y . © soroeensis = rupesins = a humilis Ta ruderanus fa 40 a
pascuorum
pomorum lapponicus © a
consobrinus
a Se as
1750 1775 1800 1825
87
1875
keriensis
asiaticus O
hypocrita
1900 1925
1950
2000
Date of first description of species
Fig. 7 Scatterplot of 239 presently accepted bumble bee species by numbers of infrasubspecific names (from a manuscript catalogue, unpublished) and date of first formal description of species. The British fauna is distinguished as filled triangles, the nearly cireumpolar fauna (B. hyperboreus, B. balteatus, B. polaris and B. lapponicus) as squares, and some species
with more infrasubspecific names are labelled individually.
species appear to be known for bumble bees than for many other groups of organisms. Most of these bum- ble bee species have been described by authors working in Europe (including European Russia). The species with the largest geographic range sizes, and particu- larly the European species with the largest ranges, have tended to be described first. The same species have also attracted the highest numbers of synonyms and subspecific names. As a group, bumble bees have an unusually high ratio of synonyms and subspecific names per species, which is otherwise known for some of the groups of larger and more colourful butterflies.
A few European authors were disproportionately prolific between 1910 and 1960 in describing finer variation at infrasubspecific rank, which now accounts for one third of all bumble bee names. Again, this more detailed effort has been largely concentrated on the earlier-described species that are more widespread within Europe (in contrast, New World bumble bees have been ignored at this level), presumably because large samples were more readily accessible to the most active authors. Determining whether this re-direction of activity towards lower nomenclatural ranks was a | logical progression in the recognition of useful taxa, a | fashion in taxonomic concepts, or in some cases merely
a less disruptive channelling of the enthusiasm of
some authors to publish more names (the ‘mihi itch’), is beyond the scope of this preliminary review.
All of these patterns in the descriptions of bumble bees must, as yet, be interpreted with caution. Much work still remains to be done on the rates of descrip- tion of taxa at different nomenclatural ranks (species, subspecies, infrasubspecies), on rates of recognition of synonymy and of changes in rank, and particularly on how this activity is partitioned among different time periods, different geographic regions, different taxo- nomic subgroups and different authors.
Fundamental to almost all analyses are taxonomic revisions and checklists of bumble bee species. A revised checklist is now overdue, because nearly half (49%) of all names for bumble bees have been pub- lished since the last synoptic checklist (Skorikov, 1922a).
Development of a revised checklist
To begin to bring a checklist up to date, a draft was made in 1980 and first circulated for comment in 1985 (Williams, 1985a). This project was developed during a more detailed study of the west Himalayan fauna (Williams, 1991) and as part of continuing work on the large fauna of China in collaboration with Wang S.-f.
88
Kallobombus
a
Median names per species
Confusibombus fo)
Thoracobombus
Bombus s.str a
0 10 20 30
P.H. WILLIAMS
Laesobombus O°
40 50 60 70
Median range size per species
Fig. 8 Scatterplot of 38 bumble bee subgenera by median numbers of all names per species (including synonyms, subspecific and infrasubspecific names; from a manuscript catalogue, unpublished) and median range size per species (number of occupied 611,000 km? grid cells world-wide). The subgenera represented in the British fauna are distinguished as filled triangles and some subgenera are labelled individually.
and Yao J. (unpub.). Some of the broader revisions that have had the greatest influence on this include works by Vogt (1909, 1911), Franklin (1913), Stephen (1957), Milliron (19706, 1971, 1973a, b), Loken (1973, 1984), Pekkarinen (1979), Reinig (1981), Wang (1982, 1987, 1988), Rasmont (1983, 1988), Thorp ef al. (1983), Labougle (1990), and especially the publications by Skorikov (1910-1938) and Tkalcti (1959-1989). In- evitably, the present checklist cannot be expected to solve all biological and nomenclatural problems, but it is hoped that by identifying some of the major prob- lems it will stimulate further research.
Acknowledgements
My grateful thanks to all who have contributed to the discussion of this and previous lists, including Donald Baker, Andreas Bertsch, Barry Bolton, Sydney Cameron, Gabriela Chavarria, Sally Corbet, Liz Day, Mick Day, Anne Divers, George Else, Kevin Gaston, Chris Humphries, Ian Kitching, Astrid Loken, Rod Macfarlane, Jim Mallet, Russell Miller, Chris O’ Toole, Antti Pekkarinen, Chris Plowright, Oliver Prys-Jones, Robert Prys-Jones, Pierre Rasmont, Malcolm Scoble, Chris Starr, Bill Stephen, Robin Thorp, Borek Tkalcii,
Dick Vane-Wright, Doug Yanega, Yao Jian and Wang Shu-fang, although they do not necessarily share the opinions expressed here. I particularly appreciate the help of Philip Tubbs, Executive Secretary to the ICZN, for advice on the application of the current Code (ICZN, 1985) to nomenclatural problems. My thanks to Julie Harvey and Lorna Mitchell of the Entomology Library (Dept of Library and Information Services, NHM) for all their help. I would also like to thank Wang Shu-fang, Chen Wei and Yao Jian for their generous hospitality during my visit to China, as well as the Dept of Botany (NHM) for funding the visit.
TAXONOMY
Bumble bees are a monophyletic group (Williams, 1985b, 1995), constituting the tribe Bombini. They may be distinguished from other bees (family Apidae) by the following diagnosis (from Williams, 1991, which includes descriptions of the characters and dis- cussion of homologies):
Bombini have the labrum at least twice as broad as long. The
CHECKLIST OF BUMBLE BEES
labrum lacks a longitudinal median ridge, although for the females it has a strong transverse basal depression. The clypeus has a transverse subapical depression and the apico- lateral corners are curved back towards the occiput. A malar area (= malar space) separates the compound eye from the base of the mandible, often by a distance greater than the breadth of the mandible at its base. The hind wings lack a jugal lobe (= anal lobe). The volsella (= lacinia) of the male genitalia is greatly enlarged and is produced apically beyond the gonostylus (= squama).
Bumble bees are large (body length 7-27 mm) robust insects. Their bodies have a dense covering of variously coloured long plumose hairs, although these are few or absent on some parts of the ventral surface of the gaster, on parts of the propodeum, on parts of of the anterior face of gastral tergum I, and on parts of the head. The sclerites are usually black, or lighter brown on the distal parts of the limbs, but are never marked with bright yellow, red or metallic (= interfer- ence) colours. The wings may be transparent (= hyaline) to strongly darkened (= infuscated), but rarely show strongly metallic reflections.
Female bumble bees have 12 antennal ‘segments’ (= scape, pedicel and 10 flagellomeres) and six visible gastral terga and sterna (abbreviated to TI-VI, SI-VI). Males have 13 antennal ‘segments’ (= scape, pedicel and 11 flagellomeres) and seven visible gastral terga and sterna (abbreviated to TI-VII, SI- VID).
Where possible, a divisive, ‘top-down’ approach to the description of bumble bee diversity has been fol- lowed, in the sense of concentrating initially on higher-rank relationships and then distinguishing pro- gressively the species groups, species and then variation within species (as opposed to beginning with de- scribed infraspecific taxa and searching ‘upwards’ for close relatives). At the rank of species, this accepts those putative species or species complexes that are supported by consistent evidence for separate status, and which can be reliably identified throughout their range for the purpose of mapping distributions. This kind of broad over-view at least has the potential to apply consistent criteria across all taxa, even though it is appreciated that not all taxa at the rank of species are necessarily of the same kind (Ackery & Vane-Wright, 1984; de Queiroz & Donoghue, 1988). Specialists will need to modify this list as further information becomes available for particular species groups.
Phylogeny, supra-specific taxa and ordering of species
From available cladistic evidence (Williams, 1991, 1995), use of Psithyrus as a genus for the social parasites separate from the remainder of the social bumble bees in Bombus can no longer be justified, so a single genus Bombus is used for all of the species of bumble bees (see the comments under the subgenus Psithyrus).
A system of subgenera has become widely used by
89
specialists who wish to label assemblages of the more closely similar species. This system is summarised with subgeneric diagnoses and keys by Richards (1968). For a review of supraspecific classifications of bumble bees, see Ito (1985).
The subgeneric system would be more useful if the names were applied only to strictly monophyletic groups. Unfortunately, Richards’s (1968) concepts of the bumble bee subgenera do not always agree well with recent estimates of phylogeny, because some of these assemblages now appear to be paraphyletic (e.g. Mendacibombus) or even polyphyletic (e.g. Sibiricobombus in the sense of Richards, 1968, in- cludes Obertobombus, whereas he placed B. (Sibiricobombus) flaviventris in Subterraneobombus) (Williams, 1991).
Furthermore, the system of subgenera would prob- ably be more useful if it were simplified (e.g. Menke & Carpenter, 1984; and reply by Williams, 1985c). For example, in the New World, both the monophyletic
fraternus-group of subgenera and the subgenus
Fervidobombus are endemic, and these are the only two groups represented south of the Panama isthmus. But whereas Fervidobombus has been treated nearly consistently as a single, relatively large subgenus (20 species in this list), the fraternus-group (18 species in this list) has regularly been split into as many as nine subgenera.
However, no attempt is made in this checklist to revise radically the subgeneric system, because stabil- ity will only be served when a revision can be supported by a comprehensive cladistic analysis. This should include not only a broad sample of species, but also a broad range of morphological and molecular charac- ters. Minor modifications from the subgeneric system described by Richards (1968) are detailed in the list after the subgeneric names.
Full synonymy of supraspecific names is included in this checklist, along with details of type species, because these have been revised since Richards (1968). The given generic combination for subgeneric names is Shown. Where a genus-group name was published at the rank of genus and subsequently treated at subgeneric rank, the first such action is listed separately. The two- letter abbreviations for subgeneric names are based on those used by Ito (1985).
Species are listed in an order (Table 3) that repre- sents their phylogenetic relationships (after the sequencing convention of Nelson, 1972) as these are currently understood from cladistic studies of the adult morphology of both sexes (Williams, 1995, and many references therein). Within subgenera, this informa- tion is still of a very preliminary nature (e.g. Williams, 1991). Many other estimates of relationship exist and would result in different sequences of species names. An alphabetic index is provided as an aid to finding names in this list.
90
Table 3 List of names for subgenera of the genus Bombus, with numbers of species recognised in this checklist. The subgeneric classification is based on Richards (1968), modified to accommodate recent publications (see text; no attempt is made to revise the subgeneric system, because stability will only be served when a revision can be supported by a comprehensive cladistic analysis). Subgenera are listed in an order that represents their phylogenetic relationships (after the sequencing convention of Nelson, 1972) as these are currently understood from cladistic studies of the adult morphology of both sexes (Williams, 1995).
Subgenus Number of species
1 Mendacibombus 12
2, Bombias 2
3 Confusibombus 1
4 Mucidobombus 1
5 Eversmannibombus 1
6 Psithyrus 29
7 Laesobombus 1
8 Orientalibombus 3
9 Exilobombus 1 10 Thoracobombus 19 11 Tricornibombus 3 12 Fervidobombus 20 13 Senexibombus 4 14 Diversobombus 4 15 Megabombus 14 16 Rhodobombus 3 17 Kallobombus 1 18 Alpinobombus 3) 19 Subterraneobombus 9 20 Alpigenobombus 6 2] Pyrobombus 43 22 Festivobombus 1 23 Rufipedibombus 2 24 Pressibombus 1 5 Bombus s.str. 10 26 Cullumanobombus 4 27 Obertobombus 2, 28 Melanobombus 14 29 Sibiricobombus 5 30 Fraternobombus 1 31 Crotchtibombus 1 32 Robustobombus 3) 33 Separatobombus 2 34 Funebribombus 2} 35 Brachycephalibombus 2 36 Rubicundobombus 1 37 Coccineobombus 2 38 Dasybombus 2
Criteria to discriminate species
It is not possible or appropriate to discuss species concepts in detail in this paper (though the selected references provide some introduction; for recent re- views, see Claridge et al., 1997; Mallet, 1997). However, in order to interpret the checklist, where possible it would be useful to make the species-dis- criminating criteria explicit. It is equally important to convey the present belief that there is no simple solu-
P.H. WILLIAMS
tion to the problem, and that no single known approach can resolve all of the cases in a uniform and entirely satisfactory manner.
Species concepts (ideas or general notions of the class of objects) and species diagnoses (operational determinations of individual objects) are contentious and probably unresolvable issues. Therefore there is arguably no single ‘true’ list of species, only more or less valid interpretations from different viewpoints.
Unresolveable conflicts may arise from opposing views of the nature of species. Species have been regarded either as typological classes, with member- ship to be defined by some shared essence (reviewed by Templeton, 1981), or as individuals, to be discoy- ered (Ghiselin, 1975). There are also conflicting opinions concerning criteria (characteristics or stand- ards by which an object may be judged) for recognising species, based in part on differing emphasis on pattern or process (de Queiroz & Donoghue, 1988).
Species may be considered not to differ from taxa at other ranks (e.g. genera, subspecies) in any qualitative way. There may be quantitative differences in the numbers of character differences that distinguish them in comparison with taxa of lower rank. For example, according to Mallet (1995:294), Darwin (1859) held this view. The problem with quantitative criteria (whether applied to genetic or phenotypic characters) is there is no reason to believe that any choice of threshold in the degree of difference used to recognise taxa at the rank of species is anything other than essentially arbitrary and thus idiosyncratic to particu- lar authors.
In another view, species may be considered to differ qualitatively from taxa at other ranks. It is widely accepted, though often implicitly, that taxa at the rank of species should be recognised so as to mark the boundary between, on the one hand, reticulate rela- tions (for sexually reproducing organisms), and on the other, more consistently divergent genealogical rela- tions. One problem is that this distinction may require predictions as to whether or not currently distinct groups of individuals are likely to show reticulate relationships again in the future.
Interbreeding and the associated genetic recombi- nation is an important part of Dobzhansky’s (1937) ‘modern synthesis’ of Mendelian genetics with Dar- win’s natural selection theory for evolution. Emphasising interbreeding as a criterion for recognis- ing species characterised what Mayr (1940, 1963) called the ‘biological’ species concept. These ideas have been modified in the recognition concept of species (Paterson, 1985). One problem with inter- breeding or mate recognition as criteria for recognising species is that direct and reliable evidence is rarely available and the results of tests under artificial condi- tions cannot necessarily be generalised (Splitter, 1982). Another is that the capacity for interbreeding is an
'
|
——eE
CHECKLIST OF BUMBLE BEES
ancestral condition (i.e. not an homology) and so cannot provide support for recognising taxa in the phylogenetic sense (Rosen, 1979).
In practice, all that is usually available to discrimi- nate species as ‘different’ is evidence from character differences and their patterns of concordance among individuals. The phylogenetic species concept (Cracraft, 1989) is popular because it also embodies the notion that species mark the boundary between different patterns of relationship among individuals and yet it does not rely on inference of interbreeding. The problem is that discovery of phylogenetic species as minimum cladistically-diagnosable (discrete) groups of individuals requires that these groups uniquely share homologies (synapomorphies), which may not always be the case (Ackery & Vane-Wright, 1984; Frost & Kluge, 1994).
Mallet (1995) has argued for minimising the number of assumptions built into species concepts. He sug- gests that two nominal taxa should be considered conspecific until it can be demonstrated that data for multiple characters distinguish consistent subgroups of individuals with few or no intermediates (the char- acter-cluster concept of species). Although he was arguing against the use of the widely-held biological species concept, he recognised that his prescription differs little from recent common practice. The prob- lem with the cluster concept is how to decide on a threshold for permissable numbers of intermediate individuals between taxa for them still to be consid- ered separate species.
Ultimately, species may be seen as useful conven- tions to aid in the communication of information gathered about the individuals that are their parts. It may be argued that the most important initial goal is to describe the nature of the variation in each particular case and to avoid presenting only theory-laden (and constrained) interpretations. In this way, basic infor- mation on variation will remain available for re-interpretation as theory changes.
For the sake of illustration, four principal classes of problems in geographical variation may be distin- guished within the spectrum of kinds of relationships, with the following examples:
Broad co-occurrence of differing individuals
Skorikov (1931) and Reinig (1935) recognised that throughout much of the range of B. keriensis, both yellow-banded and cream- or white-banded individu- als with indistinguishable morphology co-occur (Fig. 9). From available evidence, it is possible that B. niveatus / vorticosus may show a similar pattern of
_ yellow/white variation, as may B. impetuosus/potanini,
although with differing degrees of geographical varia- tion in colour-form frequency (see the comments on these species). Consequently, taxa in these pairs are
91
also treated as conspecific for the present (it is possible that in some cases such colour differences may be controlled by alleles at a single locus, see Owen & Plowright, 1980, on B. melanopygus; and Williams, 1991, on B. asiaticus; or by small numbers of loci, see Plowright & Owen, 1980, on B. rufocinctus). In con- trast, although the yellow-banded B. shaposhnikovi and the white-banded B. handlirschianus also show a broadly-overlapping pattern of distribution, the one white-banded male that I have seen is distinct from the yellow-banded males in the morphology of its genita- lia (Williams, 1991).
Broad clinal variation
Many species show broad trends in variation across continents, most obviously in colour pattern (e.g. B. cingulatus, Fig. 10; and the trifasciatus-group, Fig. 13, which may be combined with locally convergent colour variation, e.g. within the haemorrhoidalis- group, breviceps-group and rotundiceps-group, see Sakagami & Yoshikawa, 1961; Tkalcii, 19685, 1989). In North America, several pairs of nominal taxa were described originally from individual type-specimens with differing colour patterns from eastern and west- ern regions respectively (e.g. B. auricomus /nevadensis, B. fervidus / californicus, B. pensylvanicus / sonorus, B. terricola / occidentalis). These taxon pairs have long caused difficulties, for example with Franklin (1913:239) commenting on a list including these taxa and others that are now considered conspecific that “it must be entirely a matter of personal opinion whether they should be given full species rank or be considered as only subspecies’ (although, intriguingly, B. auricomus /nevadensis were not included in Franklin’s list). In at least some of these cases, many individuals with what appears to be a continuum of intermediate colour patterns are now known from broad intervening areas, so that threshold criteria for distinguishing these taxa appear to be essentially arbitrary (e.g. making decisions based on whether a particular tergum has the pubescence entirely yellow, rather than having a few black hairs present). In consequence, taxa in these taxon pairs are treated here as conspecific and maps are compiled for the more clearly recognisable, more inclusive taxa (but see the comments on B. auricomus / nevadensis).
Narrow hybrid zones
In some cases, otherwise discrete colour forms with closely similar morphology meet in narrow zones (of the order of a few km in breadth), where there may be evidence of intermediate or genetically recombinant individuals. In Europe this is best known for B. ruderatus / argillaceus (Fig. 11; Scholl, Obrecht & Zimmermann, 1992), and inAsia it has been suggested
oD P.H. WILLIAMS
| § ‘
Fig.9 Approximate distribution range (area within the dotted line) and principal colour variation for B. keriensis from Reinig (1939: fig. 23). Many more records are available now, but the pattern remains similar, with broad overlap of yellow- and white-banded individuals in Mongolia, Tien Shan, Pamir and western Himalaya. Yellow and cream pubescence is shown on the bees by crosses; red pubescence by vertical hatching.
CHECKLIST OF BUMBLE BEES 93
ombus : /uderans suderatis abr OTGUMACEUS SCOP.
Fig. 11 Distribution records (spots), approximate range (area within the cross-hatching, left, and line, right) and principal colour variation between queens of B. ruderatus and B. argillaceus in Europe from Reinig (1939: fig. 7). These taxa were regarded as subspecies by Reinig, but have recently been treated as separate species. Although there is evidence of a hybrid zone between some areas of parapatry, the hybrid individuals are very rare (Scholl, Obrecht & Zimmermann, 1992). Yellow pubescence is shown on the bees by crosses.
my
Fig. 12 Distribution records (spots) and principal colour variation for B. asiaticus in Kashmir from Williams (1991: map 48). There is evidence of a hybrid zone between some areas of parapatry, such as some high passes along the divide of the Great Himalaya Range, where there are abundant hybrid individuals. The spot symbols show the locally most abundant colour pattern. Yellow pubescence is shown on the bees by fine stippling; red pubescence by vertical hatching.
94
P.H. WILLIAMS
Fig. 13 Distribution records and principal colour variation within the trifasciatus-group in Asia (updated from Williams, 1991: fig. 11). The individuals may all be considered parts of a single species, B. trifasciatus, depending on which species-defining criterion is accepted. The dashed line shows the 1000 m contour above sea level and the solid line shows the 4000 m contour. Yellow pubescence is shown on the bees by fine stippling, orange pubescence by coarse
CHECKLIST OF BUMBLE BEES
<@
96
for B. asiaticus / longiceps (Fig. 12; Williams, 1991). Other possible examples include B. lapponicus / monticola in Europe and B. pyrosoma / friseanus / miniatus in China. In the case of B. asiaticus / longiceps, | have treated them as conspecific, because intermediate individuals greatly outnumber ‘typical’ individuals at some localities. For the other cases, I have followed earlier treat- ments of these taxa as separate species, because intermediate individuals are rare or not well known (although this may be a consequence of poor sam- pling in some inaccessible areas).
Disjunct peripheral populations
Some peripheral populations on offshore islands or habitat islands (e.g. mountains) show some diver- gence in colour pattern with little morphological divergence. European examples include B. terrestris / canariensis and B. hortorum / reinigiellus. Asian examples include B. schrencki / honshuensis, B. trifasciatus / maxwelli (Fig. 13, Peninsular Malay- sia), B. trifasciatus / wilemani (Fig. 13, Taiwan), B. breviceps / angustus, B. parthenius / sonani, B. flavescens / rufoflavus and B. flavescens / baguionensis. For the application of the biological species concept, in these cases there is no ‘natural’ meeting of individuals between the taxon pairs and so no admissible evidence on interbreeding (Splitter, 1982). For the application of Mallet’s (1995) cluster concept, quantitative analysis of patterns of variation is urgently needed. Where this information is absent, I agree with his prescription of treating taxa in these taxon pairs as provisionally conspecific. Bombus honshuensis and B. schrencki are mapped separately here because, from published accounts and a small sample of material examined, their colour differ- ences appear to coincide with stronger and more consistent morphological distinctions.
It is hoped that further information may help to clarify these cases. In the interests of pluralism, I aim to report not only a preferred interpretation in the comments on each species, but also at least the more widely-held alternative interpretations.
Sub-specific taxa
For this checklist the interest is primarily in problems of recognition and nomenclature for taxa at the rank of species. Subspecific names refer to parts of species, and so for present purposes these can be treated as synonyms of specific names (e.g. Schwarz ef al., 1996). This is not to say that subspecific taxa should not be recognised if they are considered useful, and of course other biologists may add subspecies to this list (cf. Rasmont et al., 1995).
P.H. WILLIAMS
NOMENCLATURE
Nomenclature should be seen as the servant of biol- ogy: its purpose is to provide labels that enable biologists to communicate information about organ- isms with minimal confusion concerning the organisms to which they refer. Accounts of the history of nomen- clature for many groups of organisms (e.g. on British bumble bees: Alford, 1975; Prys-Jones & Corbet, 1987:82) show that this is not a trivial matter and that rules are necessary.
Treatment of names follows the /nternational Code of Zoological Nomenclature (International Commis- sion on Zoological Nomenclature [{CZN], 1985). The Principle of Priority is generally adhered to, although regard is given to the stated purpose of priority (ICZN, 1985: Article 23b): namely that it should be used to promote stability and is not intended to be used to upset a long-accepted name in its accustomed mean- ing (Article 79c) through the introduction of an unused name that is its senior synonym. Similar action is also suggested where cases of homonymy affect current usage, although this action cannot be taken when it is felt desirable to maintain availability of a senior homo- nym. My suggestions for applications to ICZN for conservation of names in current use are indicated by
stars (@).
Typographical conventions
Bombus b-us valid name in the species group,
c-us available name in the species group, including synonyms of a valid spe- cies name,
2d-us available name in the species group,
a provisional synonym of a valid
species name,
unavailable name, informally asso-
ciated with a valid species name,
jf-us examined type material for species-group name f-us examined (in whole or in part),
[e-us]
@ comments on status of species, O comments on application of names, Ok) suggestion for application to ICZN.
?Bombus g-us_ valid name in the species group, fora taxon that is recognised provision- ally as a separate species from B.
b-us.
A question mark (?) before a valid name shows that, while it refers to a taxon that is considered likely to be a separate species, it may be conspecific with the preceding taxon in the list (1.e. while Bombus g-us may be conspecific with Bombus b-us, Bombus d-us is much more likely to be conspecific with Bombus b- US).
CHECKLIST OF BUMBLE BEES
Names in the more detailed references are followed by names of authors, date of first publication (within the meaning of ICZN, 1985), and page reference. Wherever possible, the true first date of publication is given in preference to any purported date of publica- tion when these differ. If a name were published originally in a different generic combination, then the original genus is shown in brackets. If the name had been published originally with a different termination, or with capital initial letters, diacritic marks etc., then the original form is shown without the mandatory changes (with the exception that small capital letters are reduced to lower case).
Selection of synonyms
This checklist is based on a much longer catalogue of over 2800 names. As a checklist, it is not required to include the full list of synonyms, so synonyms are selected for this list primarily where they help to clarify the identity and scope of the species (including the subspecies included by some authors), particularly with reference to those names in most common use in the literature of the last 25 years. Misidentifications are not included with the lists of synonyms and are discussed only when necessary to clarify the applica- tion of problematic names.
Applications to ICZN
Flexibility in interpretation of the status of taxa is possible where the evidence to distinguish among interpretations is absent, inconclusive, or may permit different interpretations under different species con- cepts. Otherwise flexibility in the application of names depends on whether systematists are eager to apply to the International Commission on Zoological Nomen- clature to use its Plenary Power in order to conserve a preferred usage of names (e.g. Laken et al., 1994; ICZN, 1996).
I propose that this could be achieved in a single application to include all names for which action is currently known to be required (atratus, balteatus, distinguendus, flavifrons, humilis, hyperboreus, mesomelas, mixtus, norvegicus, polaris, pyrenaeus, soroeensis and variabilis). Comments on this proposal would be welcomed.
DISTRIBUTION MAPS
This checklist was compiled in conjunction with dis- tribution data in support of biogeographic studies.
97
Maps of world-wide distribution at a coarse grain size were designed for use in comparisons of regional bumble bee faunas (e.g. Williams, 1989, 1991, 1993, 1995, 1996a, b; Williams & Seddon, 1993; Williams & Humphries, 1996).
Aside from any difficulties in identifying species or localities, comparisons among faunas are complicated by two principal factors: first, by differences in sam- pling effort (as illustrated by ‘species-accumulation curves’, e.g. Colwell & Coddington, 1994); and sec- ond, by differences in the extent of sampling areas (‘species-area effects’, e.g. Connor & McCoy, 1979). Fortunately for the first problem, the attractiveness of bumble bees to collectors has ensured that they have been relatively intensively sampled, so that most fau- nas are relatively well known. But in order to reduce this problem further, rather than extrapolate local rich- ness and lose information on individual species, the expected distributions of some species are interpo- lated on the basis of knowledge of their habitat associations (see the legend to Fig. 14). To reduce the second problem of species-area effects, equal-area grid cells were established using a cylindrical, equal- area projection of the world, marked at intervals of 10° of longitude and calculated intervals of latitude (Fig. 14). However, this does not ensure equal land areas among grid cells, or equal areas of habitat suitable for bumble bees.
Because the intention is to study biogeographic patterns, maps are required to show all historical records, including data from areas where species may now be extinct. On the other hand, data exclude fossil taxa (reviewed by Zeuner & Manning, 1976) and documented introductions (e.g. Oliff, 1895; Frison, 1925b; Gurr, 1957; Prys-Jones et al., 1981; Arretz & Macfarlane, 1982; Cardale, 1993).
The maps for every species are not included with this checklist because many data are still being col- lected, although for each subgenus a preliminary map of species richness is included as a general guide (or for monotypic subgenera, a map of records for the single species is included). The numerical values for the grey-scale classes differ between maps and are not shown. This is because I have adopted an alterna- tive approach of using equal frequency classes, which have the advantage that each grey-scale class remains consistent in its relative richness among all maps (e.g. dark grey always shows the richest one fifth of occu- pied cells excluding the maximum etc.). The sources of the distribution data have not been included be- cause this will be included in a later atlas.
98 P.H. WILLIAMS
Key to map symbols:
Maps for single species
specimens examined,
precise literature records (e.g. ‘Dungeness TRO1, UK’),
vague locality data (e.g. “Florida’),
interpolations of expected distribution (following common practice for range-filling maps; the rules adopted here are to fill cells between occupied cells when filled cells are known to have had a high proportion of suitable habitat within recorded history; these records amount to < 10% of all gridcell records at this scale, Williams, 1993).
OOS®e
Maps for multiple species
maximum species counts are shown in black, otherwise counts are divided into five grey-scale classes of approximately equal size by numbers of grid cells.
Fig. 14 Map of the world (excluding Antarctica) using a cylindrical equal-area projection that is orthomorphic (minimum shape distortion) at 46° North and South (where bumble bee records are particularly plentiful). Intervals of 10° longitude (top of map) are used to calculate intervals of latitude (right of map) that provide equal-area grid cells of c. 611,000 km?. The portion of the grid shown covers the known, native distribution of bumble bees. Map symbols are shown above for (a) plotting individual species, for which different spots distinguish different data categories (Map 3); or (b) for plotting coincidence maps for multiple species, using a grey scale for variation in species richness (Map 1).
CHECKLIST OF BUMBLE BEES
(plot of total species richness with grey scale, for explanation see Fig. 14)
Genus BOMBUS Latreille in the broad sense
[Bremus [Jurine], 1801:164, type-species Apis terrestris Linnaeus (= Bombus terrestris (Linnaeus)) by subse- quent designation of Morice & Durrant, 1915:429, suppressed by ICZN, 1939]
Bombus Latreille, 1802a:437, type-species Apis terrestris Linnaeus (cited asApis terrestris F.) (=Bombus terrestris (Linnaeus)) by monotypy
Bombus Latreille, 1802b:385, type-species Apis terrestris Linnaeus (=Bombus terrestris (Linnaeus)) by monotypy, redescribed
[Bremus Panzer, 1805:pl. 19-21, type-speciesApis agrorum Fabricius (= Bombus pascuorum (Scopoli)) by subse- quent designation of Sandhouse, 1943:532, suppressed by ICZN, 1954]
[Bombellus WE, 1931:248, incorrect subsequent spelling]
Subgenus MENDACIBOMBUS Skorikov Mendacibombus Skorikov, 1914a:125, type-species Bombus mendax Gerstaecker by subsequent designa- tion of Sandhouse, 1943:572 Bombus (Mendacibombus) Kriiger, 1917:62
COMMENT. The species of Mendacibombus appear to be paraphyletic with respect to the rest of the
oY)
bumble bees and in consequence are not a ‘natural’ group (Williams, 1991, 1995).
Bombus (Md.) avinoviellus (Skorikov) avinoviellus (Skorikov, 1914a:126 [Mendacibombus]) ex- amined callophenax Cockerell, 1917:122, examined
Bombus (Md.) mendax Gerstaecker mendax Gerstaecker, 1869:323, examined latofasciatus Vogt, 1909:50, not of Vogt, 1909:42 (= B. lucorum (Linnaeus)) pyrenes (Tkalcit, 1975:173 [Mendacibombus]) replace- ment name for latofasciatus Vogt, 1909:50
Bombus (Md.) makarjini Skorikov makarjini Skorikoy, 1910a:329, examined
Bombus (Md.) superbus (Tkalcit)
superbus (Tkalctt, 1968a:22 |Mendacibombus]) examined
Bombus (Md.) himalayanus (Skorikov)
?varius (Skorikoy, 1914a:125 [Mendacibombus]) exam- ined, not of Lepeletier, 1832:381 (= B. campestris (Panzer))
himalayanus (Skorikov, 1914a:127 [Mendacibombus}) examined
Bombus (Md.) marussinus Skorikov marussinus Skorikov, 1910a:330, examined afghanus Reinig, 1940:230, examined
Bombus (Md.) turkestanicus Skorikov turkestanicus Skorikoy, 1910a:329, examined
Bombus (Md.) defector Skorikov defector Skorikov, 1910a:330 ?altaicus Skorikov, 1910a:329, not of Eversmann, 1846:436 (= B. melanurus Lepeletier) ?margreiteri Vogt, in Skorikov, 1910a:330, examined
@ TAXONOMIC sTATUS. Skorikov’s (1910a) de- scriptions of varieties of B. mendax are all of females. Many of these nominal taxa have subsequently been treated as separate species (e.g. Skorikov, 1931; Rasmont, 1988).
However, I have examined type material or other material identified by Skorikov for all of these taxa and find some of them to be morphologically closely similar. The females of defector, altaicus and margreiteri differ from one another principally in
100
colour, and the only males I have seen associated with them (collections in London, Petersburg, Beijing) have very similar genitalia (which are distinct from B. mendax).
Until more evidence to the contrary is available from critical studies of patterns of variation, I shall continue to treat B. defector, B. altaicus and B. margreiteri as parts of a single variable species, B. defector (Williams, 1985a, 1991).
O NOMENCLATURE. Williams (1991) regarded B. defector, B. altaicus and B. margreiteri as likely to be conspecific and following the Principle of First Re- viser (ICZN, 1985: Article 24) chose B. defector as the name for the species.
Bombus (Md.) handlirschianus Vogt Handlirschianus Vogt, 1909:49
Bombus (Md.) shaposhnikovi Skorikov shaposhnikovi Skorikoy, 1910a:329
Bombus (Md.) waltoni Cockerell chinensis Skorikov, 1910a:330, examined, not of Morawitz, 1890:352 (= B. chinensis (Morawitz)) waltoni Cockerell, 1910b:239, examined
Bombus (Md.) convexus Wang lugubris Morawitz, 1880:339, examined, not of Kriechbaumer, 1870:159 (= B. maxillosus Klug) convexus Wang, 1979:190, examined
Subgenus BOMBIAS Robertson
Bombias Robertson, 1903:176, type-species Bombias auricomus Robertson (?= Bombus nevadensis Cresson) by original designation
Bombus (Bombias) Franklin, 1913:138
Nevadensibombus Skorikoy, 1922a:149, type-species Bombus nevadensis Cresson by subsequent designation of Frison, 1927:64
Bremus (Boopobombus) Frison, 1927:59 (proposed as a
P.H. WILLIAMS
section name but stated by Frison to include those forms considered by Franklin, 1913, to belong to the subgenus Bombias Robertson), type-species Bombias auricomus Robertson (= Bombus auricomus (Robertson)) by sub- sequent designation of Williams, 1995:339.
Bombus (Bi.) nevadensis Cresson nevadensis Cresson, 1874:102
COMMENT. A single queen of B. nevadensis has been reported from Hidalgo, Mexico, by Milliron (1971) and Hurd (1979), although the species is not listed for Mexico by Labougle (1990).
?Bombus (Bi.) auricomus (Robertson) auricomus (Robertson, 1903:176 [Bombias])
@ TAXONOMIC STATUS. B. nevadensis and B. auricomus have been regarded both as conspecific (e.g. LaBerge & Webb, 1962; Milliron, 1971; Thorp er al., 1983; Laverty & Harder, 1988) and as separate species (e.g. Franklin, 1913; Rasmont, 1988; Scholl, Thorp, Owen & Obrecht, 1992; Poole, 1996).
B. nevadensis from western North America was not mentioned in the original description of B. auricomus (lectotype worker from Illinois by designation of Milliron, 1971:78), although the latter was described using characters of morphology and of colour pattern. The two taxa have generally been distinguished on the basis of the extent of the black pubescence on the dorsum of the female thorax and laterally on the male gastral terga (e.g. Franklin, 1913).
The only study to investigate variation in characters used to distinguish the two taxa at a fine spatial scale in their area of overlap was by LaBerge & Webb (1962). They reported (p. 26) that “Throughout the broad middle half of Nebraska nevadensis seems to be rather rare and most specimens, although referable to sub- species auricomus show some indication of intergrading with the typical subspecies [nevadensis] in the west. . .. Many specimens from Nebraska in the range of the typical subspecies [nevadensis| show some tendency toward the darker coloration of subspe- cies auricomus. They concluded that these variable bees are all parts of the same species.
Recently, Scholl et al. (1992) distinguished two groups of individuals on the basis of differing mobility morphs of five enzymes. The individuals in one en- zyme group were all extensively dark-banded, and Scholl et al. associated these with the name B. auricomus. However, individuals in the other enzyme group, which Scholl et al. associated with the name B. nevadensis, apparently included not only the contrast- ing, extensively pale individuals (B. nevadensis), but also a few of the extensively dark-banded individuals (B. auricomus) similar to those in the first group (8/49
CHECKLIST OF BUMBLE BEES
individuals had gastral tergum I almost completely black; 3/49 individuals had the scutellum predomi- nantly black). Thus the enzyme evidence does identify two groups of individuals, but (1) these do not appear to correspond precisely to the two traditional colour groups; (2) some of the key areas likely to support intermediate or recombinant individuals still need to be sampled for enzyme variation (e.g. in the Dakotas, L. Day in litt.); and (3) inheritance of enzyme and colour states needs to be better understood, including the unusual enzyme morphs of the heterozygous bees (detected in 20/141 queens). They concluded that these bees represent two species.
A. Scholl (in litt.) reports a further intriguing morphometric study. A random subsample of 20 queens from the enzyme study was scored for 15 characters and analysed by linear discriminant analysis. This method seeks a combination of characters that best discriminates any two a priori sets, in this case using three measurements of parts of the radial cell, eye and antenna. However, although this approach may be useful for discriminating previously recognised taxa, it does not provide evidence that they are necessarily separate species (it could also be used to discriminate morphological subsets within a single, variable popu- lation, e.g. among breeds of domestic dogs).
From an examination of 41 females, so far I have found only one subtle morphological character to distinguish eastern, banded bees (B. auricomus), on the one hand, from western unbanded (B. nevadensis) and banded (e.g. Vancouver Island) bees, on the other. This concerns the anterior part of a band of large punctures along the inner eye margin, dorsally oppo- site the ocelli, just before these punctures meet a more anterior, very dense patch of small punctures. The western bees have areas between the large punctures conspicuously shining, with few fine punctures and lacking microsculpture. In contrast, the eastern bees have these areas appearing rather dull, often with more of the fine punctures, and more particularly with a very fine, wrinkled or reticulate microsculpture. A similar difference may be present in the males, posterio- laterally to the ocelli, though the sample sizes available to me are too small for much confidence.
I regard the conflicting evidence available at present as not entirely conclusive as to whether these bees are parts of the same population or two separate species. As far as is known, both the variations of the colour pattern and of the enzyme mobilities are inherited and genetically determined, but details of patterns of in- heritance and of the spatial aspects of any association between these characters are unknown. In view of the multiple enzymes differences found and of the appar- ent association between the enzyme groups and the morphological character states, I shall follow the treat- ment of these taxa as two separate species until more evidence is available.
(plot of records for a single species, for explanation and key see Fig. 14)
Subgenus CONFUSIBOMBUS Ball
Bombus (Confusibombus) Ball, 1914:78, type-species Bombus confusus Schenck by monotypy
Bombus (Sulcobombus) Kriiger, 1917:65, type-species Bombus confusus Schenck by subsequent designation of Sandhouse, 1943:602
Confusobombus Skorikoy, 1922a:156, type-species Bombus confusus Schenck by subsequent designation of Richards, 1968:214
Bombus (Cf.) confusus Schenck confusus Schenck, 1859:135 paradoxus Dalla Torre, 1882:18 festivus Hoffer, 1882:80, not of Smith, 1861:152 (= B. festivus Smith)
@ TAXONOMIC STATUS. B. confusus and B. paradoxus differ in the colour pattern of the pubes- cence (e.g. Reinig, 1939: fig. 19). Rasmont (1988) reports that in north western Europe, the yellow-banded and white-tailed B. paradoxus occurs only as rare individuals within the population of predominantly unbanded and red-tailed B. confusus. In contrast, all of the individuals that I have seen from the disjunct population in Central Asia have the yellow-banded and white-tailed B. paradoxus colour pattern.
102
Subgenus EVERSMANNIBOMBUS Skorikov Agribombus (Eversmannibombus) Skorikoy, 1938a:145, type-species Mucidobombus eversmanniellus (=
Bombus persicus Radoszkowski) by monotypy
Bombus (Eversmannibombus) Richards, 1968:214
Bombus (Ey.) persicus Radoszkowski
calidus Eversmann, 1852:133, examined, not of Erichson in Middendorff, 1851:65 (= B. hypnorum (Linnaeus))
persicus Radoszkowski, 1881:v, examined
Persicus Radoszkowski, 1883:214, redescribed
eversmanni Friese, 1911:572, not of Skorikov, 1910¢:581 ( B. modestus Eversmann), replacement name for calidus Eversmann, 1852:133
eversmanniellus (Skorikoy, 1922a:149 [Mucidobombus]) replacement name for eversmanni Friese, 1911:572
Subgenus PSITHYRUS Lepeletier Psithyrus Lepeletier, 1832:373, type-species Apis rupestris Fabricius (= Bombus rupestris (Fabricius)) by subse- quent designation of Sandhouse, 1943:572 Apathus Newman, 1835:404, replacement name for Psithyrus Lepeletier, incorrectly stated to be a junior homonym of Psithyros Hiibner, [1819]:132 (= Macroglossum Scopoli, 1777:414)
P.H. WILLIAMS
?Psithyrus (Laboriopsithyrus) Frison, 1927:69, type-spe- cies Bombus laboriosus Fabricius (= Emphoropsis laboriosus (Fabricius) in the sense of Frison (=Bombus citrinus (Smith), a misidentification, see Milliron, 1960:99, requiring designation by ICZN) by original fixation &
Psithyrus (Ashtonipsithyrus) Frison, 1927:69, type-spe- cies Apathus ashtoni Cresson (= Bombus ashtoni (Cresson)) by original designation
Psithyrus (Fernaldaepsithyrus ) Frison, 1927:70, type-spe- cies Psithyrus fernaldae Franklin (= Bombus fernaldae (Franklin)) by original designation
Psithyrus (Eopsithyrus) Popov, 1931:134, type-species Apathus tibetanus Morawitz (= Bombus tibetanus (Morawitz)) by original designation
Psithyrus (Metapsithyrus) Popov, 1931:135, type-species Apis campestris Panzer (= Bombus campestris (Pan- zer)) by original designation
Psithyrus (Allopsithyrus) Popov, 1931:136, type-species Apis barbutella Kirby (= Bombus barbutellus (Kirby)) by original designation
Psithyrus (Ceratopsithyrus) Pittioni, 1949:270, type-spe- cies Psithyrus klapperichi Pittioni (= Bombus cornutus (Frison)) by original designation
Psithyrus (Citrinopsithyrus ) Thorp inThorp et al., 1983:50, type-speciesApathus citrinus Smith (=Bombus citrinus (Smith)) by original designation
Bombus (Psithyrus) Williams, 1991:44
[Psithyrus (Fernaldepsithyrus) Amiet, 1996:86, incorrect subsequent spelling]
@ TAXONOMIC sTATUS. It has long been consid- ered useful to regard Psithyrus as a separate genus in recognition of the distinctive behaviour of the species, as social parasites in colonies of the remaining Bombini, and in recognition of their distinctive mor- phology. However, most recent studies have shown (if phenograms are interpreted along with cladograms as phylogenetic estimates) that, although Psithyrus is itself very likely to be monophyletic, the remaining bumble bees are not (Plowright & Stephen, 1973; Obrecht & Scholl, 1981; Ito, 1985: Williams, 1985), 1991, 1995; Pamilo et al., 1987).
I have previously attempted to retain the use of the names Psithyrus and Bombus for monophyletic genera by recognising a third genus, Mendacibombus (Williams, 1985b). However, further study of all of the species of Mendacibombus (Williams, 1991, 1995) showed that it is likely to be paraphyletic with respect to all other bumble bees, with the consequence that as many as another nine genera (mostly for single spe- cies) might be required to maintain monophyly alongside a genus Psithyrus. In the face of this evi- dence, a pragmatic solution was recommended, recognising a single genus Bombus for all bumble bees, to include Psithyrus as a subgenus. This is a return to an emphasis of the more widely shared characters and the more distant affinities for the ge- neric concept, encouraged by the opinion of Michener (1990) that bumble bees are ‘morphologically mo-
CHECKLIST OF BUMBLE BEES
notonous’ in comparison with variation among species within closely related groups such as Euglossini (or- chid bees) and Meliponini (stingless bees). One advantage of a single genus for all bumble bees is that it recognises a group for which evidence of monophyly is particularly strong, so that nomenclature is most likely to remain stable in the future. Use of a single genus Bombus for all bumble bees (Williams, 1991) has now been accepted by most recent authors (e.g. Rasmont & Adamski, 1995; Rasmont ef al., 1995; Schwarz et al., 1996).
The subgenera within the former genus Psithyrus have often been considered less distinct from one another than have the other subgenera of Bombus (Pittioni, 1939a; Ito, 1985; Williams, 19855; Michener, 1990) and therefore may be treated as synonyms of Psithyrus (Milliron, 1961; Williams, 1991, 1995). In an alternative treatment, Rasmont et al. (1995) include the former subgenera of the former genus Psithyrus as separate subgenera within the genus Bombus.
O NOMENCLATURE. The names of six species of the subgenus Psithyrus from Kashmir were explicitly stated to be new combinations with the genus Bombus by Williams (1991). Rasmont ef al. (1995) have since listed the other European species in this combination. No formal statements of new combination are made here for the remaining species of the subgenus Psithyrus because a principle of implied combinations (Poole, 1996) is followed after the change in status of Psithyrus from genus to a subgenus of Bombus.
APPLICATION TO ICZN. Because the type spe- cies of Laboriopsithyrus was misidentified (discussed by Milliron, 1960:99), ICZN is required to designate as type species whichever species will best serve nomenclatural stability (ICZN, 1985: Art. 70b). It is suggested that, in the interests of stability (ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to select the species actually involved (Bombus laboriosus in the sense of Frison, = Bombus citrinus (Smith)), which was wrongly named in the type fixation (ICZN, 1985: Art. 70b(i)).
COMMENT. The highest richness of species of the subgenus Psithyrus occurs in the Old World (there are no species known from south of Panama), al- though the earliest-diverging species appear to be North American (unpublished). This is the opposite pattern to that shown by species of the largest subgenus, Pyrobombus (see the comments on the subgenus Pyrobombus).
All species of the subgenus Psithyrus are believed to be obligate social parasites in colonies of other Bombus species (reviewed by Alford, 1975; Fisher, 1987). There is variation in the degree of host specificity. See also the comments on B. inexspectatus and B. hyperboreus.
103
Bombus (Ps.) insularis (Smith)
interruptus Greene, 1858:11, not of Lepeletier, 1832:381 (= B. rupestris (Fabricius))
insularis (Smith, 1861:155 [Apathus]) examined
consultus (Franklin, 1913:459 [Psithyrus])
? bicolor (Franklin, 1913:460 [Psithyrus]) not of H6ppner, 1897:33 (=B. soroeensis (Fabricius)) (provisional syno- nym)
crawfordi (Franklin, 1913:464 [Psithyrus])
@ TAXONOMIC stTaTUS. According to D. Yanega (in litt.), who has examined the type material, B. bicolor Franklin is conspecific with B. interruptus.
Bombus (Ps.) citrinus (Smith) citrinus (Smith, 1854:385 [Apathus]) examined contiguus (Cresson, 1863:112 [Apathus])
Bombus (Ps.) variabilis (Cresson) ®
intrudens (Smith, 1861:154 [Apathus]) examined
variabilis (Cresson, 1872:284 [Apathus]) new synonym
?guatemalensis (Cockerell, 1912:21 [Psithyrus}) (provi- sional synonym)
?sololensis (Franklin, 1915:173 [Psithyrus]) (provisional synonym)
?mysticus (Frison, 1925a:138 [Psithyrus]) (provisional synonym)
@ TAXONOMIC STATUS. Specimens in the NHM collection from Mexico and Guatemala labelled ‘intrudens’ and ‘sololensis’ appear to me to be closely similar to B. variabilis. Frison (1925a) believed that B. sololensis is acolour form of B. guatemalensis. Never- theless, he proceeded to distinguish B. mysticus as a separate species on the basis of colour pattern alone. I am unaware of any reason (other than minor differ- ences in colour pattern) why B. variabilis, B. intrudens, B. sololensis, or B. guatemalensis and B. mysticus (judging from the published descriptions at least), should not be considered conspecific.
O NOMENCLATURE. A female in the NHM collec- tion has three labels “Apathus / intrudens / Smith.’, *58.135 MEX. / (Oajaca.)’, “Holo- / type’ and I am unaware of any problems with this designation. If this is correct and the type is conspecific with B. variabilis, then B. intrudens is the oldest available name for this species. D. Yanega (in /itt.) agrees with this interpreta- tion.
€ APPLICATION TOICZN. Although B. intrudens is the oldest available name for the present interpretation of this species, the name B. variabilis has been in common use for the species since 1947 (e.g. Stevens, 1948; Chandler, 1950; LaBerge & Webb, 1962; Mitchell, 1962; Medler & Carney, 1963; Hobbs, 1966; Plowright & Stephen, 1973; Hurd, 1979; Husband et al., 1980; Michener, 1990; Poole, 1996). I know of no publications using the name B. intrudens since 1947. It
104
is suggested that, in the interests of stability (ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79) (see the comments on B. muscorum). However, the consequence of this action would be that B. intrudens would no longer be available for a species or for a subspecies of B. variabilis (Cresson).
Bombus (Ps.) suckleyi Greene Suckleyi Greene, 1860:169
Bombus (Ps.) vestalis (Geoffroy) veftalis (Geoffroy in Fourcroy, 1785[see Hagen 1862:246]:450 [Apis])
Bombus (Ps.) perezi (Schulthess-Rechberg) perezi (Schulthess-Rechberg, 1886:275 [Psithyrus])
Bombus (Ps.) ashtoni (Cresson) Ashtoni (Cresson, 1864:42 [Apathus])
Bombus (Ps.) bohemicus Seidl
nemorum (Fabricius, 1775:380 [Apis]) examined, not of Scopoli, 1763:307 (= B. subterraneus (Linnaeus)), not of Fabricius, 1775:382 (= B. distinguendus Morawitz)
bohemicus Seidl, 1837:73
?chinganicus (Reinig, 1936:8 [Psithyrus]) (provisional synonym)
hedini (Bischoff, 1936:26 [Psithyrus]) not of Bischoff, 1936:15 (= B. hedini Bischoff)
@ TAXONOMIC sTATUS. Iam unaware of any rea- son (other than the small body size of the holotype female and three paratype females of B. chinganicus) why B. bohemicus and B. chinganicus should not be considered conspecific. Consistent with this, body sizes do appear to vary considerably within British species of the subgenus Psithyrus, including B. bohemicus.
Bombus (Ps.) coreanus (Yasumatsu) coreanus (Yasumatsu, 1934:399 [Psithyrus])
Bombus (Ps.) barbutellus (Kirby) Barbutella (Kirby, 1802:343 [Apis]) examined ?richardsi (Popov, 1931:150,190 [Psithyrus]) not of Frison, 1930:6 (= B. rufipes Lepeletier) 2icenti (Maa, 1948:34 [Psithyrus]) examined
O NOMENCLATURE. Loken (1984) interpreted B. saltuum (Panzer, 1801) as conspecific with B. barbutellus. Consequently, B. saltuum would appear to be the oldest available name for this species. How- ever, Lgken made no further comment on this and used the name Psithyrus barbutellus (= B. barbutellus), possibly because she remained unsure of the identity of B. saltuum. In contrast, Warncke (1986) interpreted
P.H. WILLIAMS
B. saltuum as conspecific with B. subterraneus. See the comments on B. subterraneus.
?Bombus (Ps.) maxillosus Klug maxillosus Klug in Germar, 1817:269 lugubris (Kriechbaumer, 1870:159 [Psithyrus]) unicolor (Kriechbaumer, 1870:159 [Psithyrus]) mixta (Kriechbaumer, 1870:160 [Psithyrus]) Susterai (May, 1944:267 [Psithyrus]) not infrasubspecific after Tkalci, 1977:224
@ TAXONOMIC STATUS. As Rasmont (1988) notes, B. maxillosus 1s closely similar to B. barbutellus in morphology and habitat, so that specimens cannot always be distinguished reliably. Consequently these nominal taxa might be considered conspecific. More evidence is awaited.
Bombus (Ps.) cornutus (Frison) cornutus (Frison, 1933:338 [Psithyrus]) pyramideus (Maa, 1948:19 [Psithyrus]) examined acutisquameus (Maa, 1948:21 [Psithyrus]) examined Klapperichi (Pittioni, 1949:273 [Psithyrus]) examined, not of Pittioni, 1949:266 (= B. picipes Richards) ?canus (Tkalcit, 1989:42 [Psithyrus])
Bombus (Ps.) expolitus Tkalcit expolitus (Tkalct, 1989:44 [Psithyrus]) examined
Bombus (Ps.) turneri (Richards) turneri (Richards, 1929a:141 [Psithyrus]) examined ?monozonus (Friese, 1931:304 [Psithyrus]) not of Friese, 1909:674 (= B. lucorum (Linnaeus)) ?decoomani (Maa, 1948:26 [Psithyrus]) examined ?martensi (Tkalct, 1974b:314 [Psithyrus]) (provisional synonym)
@ TAXONOMIC sTATUS. Several of these nominal taxa have been treated as separate species. However, aside from differences in colour pattern, they are closely similar in morphology. Until more evidence to the contrary is available from critical studies of pat- terns of variation, I shall treat them as parts of a single variable species.
Bombus (Ps.) tibetanus (Morawitz) tibetanus (Morawitz, 1886:202 [Apathus]) ?latefasciatus (Friese, 1931:304 [Psithyrus])
Bombus (Ps.) chinensis (Morawitz) chinensis (Morawitz, 1890[April 30]:352 [Apathus]) morawitzi (Friese, 1905:516 [Psithyrus]) not of Radoszkowski, 1876:101 (= B. morawitzi Rado- szkowski) hénei (Bischoff, 1936:26 [Psithyrus]) not of Bischoff, 1936:10 (= B. friseanus Skorikoy)
CHECKLIST OF BUMBLE BEES
Bombus (Ps.) novus (Frison) novus (Frison, 1933:340 [Psithyrus]) ?nepalensis (Tkalcti, 19746:318 [Psithyrus]) examined
Bombus (Ps.) branickii (Radoszkowski) Branickii (Radoszkowski, 1893:241 [Psithyrus]) exam- ined chloronotus (Morawitz, 1894:6 [Apathus]) elisabethae (Reinig, 1940:231 [Psithyrus]) examined [branichi (Kim & Ito, 1987:32 [Psithyrus]) incorrect sub- sequent spelling]
Bombus (Ps.) rupestris (Fabricius) rupeftris (Fabricius, 1793:320 [Apis]) Pyrencus (Lepeletier, 1832:375 [Psithyrus]) Interruptus (Lepeletier, 1832:381 [Psithyrus]) armeniacus (Reinig, 1970:77 [Psithyrus]) not of Radoszkowski, 1877b:202 (= B. armeniacus Radoszkowski)
Bombus (Ps.) ferganicus (Radoszkowski) ferganicus (Radoszkowski, 1893:241 [Psithyrus]) exam- ined ochraceus (Morawitz, 1894:5 [Apathus]) indicus (Richards, 1929a:139) examined
Bombus (Ps.) morawitzianus (Popov) morawitzianus (Popov, 1931:148,183 [Psithyrus]) exam- ined redikorzevi (Popov, 1931:160,181 [Psithyrus])
O NOMENCLATURE. Griitte (1937) regarded B. morawitzianus and B. redikorzevi as conspecific and, following the Principle of First Reviser (ICZN, 1985: Article 24), chose B. morawitzianus as the name for the species.
Bombus (Ps.) campestris (Panzer)
campestris (Panzer, 1801(74):11 [Apis])
Varius (Lepeletier, 1832:381 [Psithyrus])
flavus (Pérez, 1884:265 [Psithyrus])
flavo-thoracicus (Hoffer, 1889:49 [Psithyrus])
?Susterai (Tkalcti, 1959:251 [Psithyrus]) examined, not of May, 1944:267 (= B. maxillosus Klug) (provisional synonym)
?susteraianus (Tkalcu, 1977:224 [Psithyrus]) replacement name for susterai Tkalcii, 1959:251 (provisional syno- nym)
@ TAXONOMIC stTaTUS. Iam unaware of any rea- son (other than minor differences) why B. campestris and B. susteraianus should not be considered conspecific.
105
Bombus (Ps.) bellardii (Gribodo) Bellardti (Gribodo, 1892:108 [Psithyrus]) examined pieli (Maa, 1948:29 [Psithyrus]) examined, new synonym tajushanensis (Pittioni, 1949:277 [Psithyrus]) examined, not of Pittioni, 1949:244 (= B. kulingensis Cockerell), new synonym
@ TAXONOMIC sTATUS. B. bellardii, B. pieli and B. tajushanensis are closely similar in morphology and I am unaware of any reason why these nominal taxa should not be considered conspecific.
O NOMENCLATURE. For this species, the oldest available name is B. bellardii, which becomes the valid name. The only subsequent publications using the name B. pieli of which I am aware are by Maa (1948), Sakagami (1972), Tkalciit (1987) and Williams (1991), so this change of valid name is not a serious disruption of common usage.
Bombus (Ps.) norvegicus (Sparre-Schneider)® norvegicus (Sparre-Schneider, 1918:40 [Psithyrus]) not of Friese, 1911:571 (= B. monticola Smith) transhaicalicus (Popov, 1927:269 [Psithyrus])
O NOMENCLATURE. With Psithyrus regarded as being a subgenus of the genus Bombus (Williams, 1991, 1995), P. norvegicus Sparre-Schneider (1918) becomes a junior secondary homonym in Bombus of B. lapponicus var. norvegicus Friese (1911) (deemed subspecific, see ICZN, 1985: Article 45g(ii)), and therefore the name P. norvegicus Sparre-Schneider is invalid (ICZN, 1985: Article 57c). For this species, the oldest available name of which I am aware is P. norvegicus var. transbaicalicus Popov, 1927 (deemed to be subspecific, see ICZN, 1985: Article 45g(ii)), so B. transbaicalicus would become the valid name.
© APPLICATION TO ICZN. Although B. trans- baicalicus is the oldest available name for this species, the name B. norvegicus has been in com- mon use for the species since 1947 (e.g. Faester & Hammer, 1970; Delmas, 1976; Ito & Tadauchi, 1981; Pekkarinen ef al., 1981; Reinig, 1981; Loken & Framstad, 1983; Rasmont, 1983; Loken, 1984; Ito, 1985; Pekkarinen & Teras, 1993; Rasmont ef al., 1995). It is suggested that, in the interests of stabil- ity, an application be made to ICZN to use its Plenary Power to suppress the senior homonym (ICZN, 1985: Article 79) (see the comments on B. muscorum). However, the consequence of this action would be that norvegicus Friese would no longer be available for a subspecies of B. monticola.
Bombus (Ps.) fernaldae (Franklin) fernalde (Franklin, 1911:164 [Psithyrus]) examined
106
Bombus (Ps.) flavidus Eversmann flavidus Eversmann, 1852:131 lissonurus (Thomson, 1872:49 [Apathus])
@ TAXONOMIC STATUS. Rasmont (1988) reports that the Pyrenean population of B. flavidus is morphometrically distinct from the disjunct Scan- dinavian population (comparable distinctions are not known within its close relatives B. norvegicus and B. sylvestris, which share these areas of distribtuion). Nevertheless he continues to treat them as conspecific and I shall follow this, at least until further evidence in support of two separate species is available.
Bombus (Ps.) skorikovi (Popov) skorikovi (Popov, 1927:267 [Psithyrus]) examined ?gansuensis (Popov, 1931:202 [Psithyrus]) ?kuant (Tkalcu, 1961b:362 [Psithyrus])
Bombus (Ps.) quadricolor (Lepeletier) Quadricolor (Lepeletier, 1832:376 [Psithyrus]) globosus (Eversmann, 1852:126 [Psithyrus]) meridionalis (Richards, 1928b:351 [Psithyrus]) not of Dalla Torre, 1879:13 (= B. hortorum (Linnaeus))
Bombus (Ps.) sylvestris (Lepeletier)
Sylvestris (Lepeletier, 1832:377 [Psithyrus])
Brasiliensis (Smith, 1854:385 [Apathus]) examined, not of Lepeletier, 1836:470 (= B. brasiliensis Lepeletier)
citrinus (Schmiedeknecht, 1883[see Baker, 1996c:297]:23[407] [Psithyrus]) not of Smith, 1854:385 (= B. citrinus (Smith))
[silvestris (Dalla Torre, 1896:571 [Psithyrus]) incorrect subsequent spelling]
e00e000ee0e er Pats eecsece: * ee se 00% oe one f
$7
Subgenus LAESOBOMBUS Kriiger Bombus (Laesobombus) Kriiger, 1920:350, type-species Bombus laesus Morawitz by monotypy Agrobombus (Laesobombus) Skorikov, 1922b:20, type- species Bombus laesus Morawitz by monotypy Agribombus (Laesibombus) Skorikoy, 1938a:145, unjusti- fied emendation
P.H. WILLIAMS
Bombus (Ls.) laesus Morawitz laesus Morawitz in Fedtschenko, 1875:3 Mocsaryi Kriechbaumer, 1877:253 ?maculidorsis (Skorikoy, 1922b:23 [Agrobombus]) not infrasubspecific after Panfilov, 1956:1328 ?tianschanicus Panfilov, 1956:1327 (provisional synonym) ferrugifer Reinig, 1971:158
@ TAXONOMIC STATUS. Panfilov (1956) regarded B. laesus, B. mocsaryi, B. maculidorsis and B. tian- schanicus as separate species, differing particularly in: (1) the colour of the pubescence on the thoracic dor- sum; (2) the number of large punctures on the clypeus; (3) the strength of the median keel on gastral sternum VI; and (4) the length of the hair of the dorsum. However, from the material I have examined (collec- tions in London, Beijing), these character states do not appear to be either discreet or strongly associated. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Subgenus ORIENTALIBOMBUS Richards Bombus (Orientalibombus) Richards, 1929c:378, type- species Bombus orientalis Smith (= Bombus haemorrhoidalis Smith) by original designation Bombus (Orientalobombus) Kruseman, 1952:102, unjus- tified emendation
Bombus (Or.) funerarius Smith funerarius Smith, 1852b:47, examined priscus (Frison, 1935:349 [Bremus]) birmanus (Tkalct, 1989:47 [Orientalibombus]) examined
Bombus (Or.) braccatus Friese braccatus Friese, 1905:512, examined metcalfi (Frison, 1935:357 [Bremus]) examined
Bombus (Or.) haemorrhoidalis Smith heemorrhoidalis Smith, 1852a:43 orientalis Smith, 1854:402, examined assamensis Bingham, 1897:550, examined
CHECKLIST OF BUMBLE BEES
montivolans Richards, 1929c:382, examined semialbopleuralis (Tkalcit, 1974b:322 [Orientalibombus]) cinnameus (Tkalcii, 1989:47 [Orientalibombus]) examined
@ TAXONOMIC sTATUS. Several of these nominal taxa have been treated as separate species, most recently in the case of B. montivolans [Burma to southern China] (e.g. Tkalcii, 1968b, 1989). However, aside from differ- ences in colour pattern, they are all closely similar in morphology with a range of variation (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Subgenus EXILOBOMBUS Skorikov Mucidobombus (Exilobombus) Skorikov, 1922a:150, type- species Mucidobombus exil Skorikoy (cited as exiln.) (= Bombus exil (Skorikoy)) by monotypy Megabombus (Exilnobombus) Milliron, 1973a:81, unjus- tified emendation
Bombus (Ex.) exil (Skorikov)
exiln. nov. (Skorikovy, 1922a:150 |[Mucidobombus}) {not a replacement name]
[exul (Skorikov, 1931:216 [Mucidobombus}) incorrect sub- sequent spelling]
exil (Milliron, 1961:56 [Megabombus]) justified emenda- tion
[exilis Richards, 1968:254, incorrect subsequent spelling]
exul (Tkalcti, 1974a:42 [Megabombus]) unjustified emen- dation
107
species Apis sylvarum Linnaeus (= Bombus sylvarum (Linnaeus)) by subsequent designation of Sandhouse, 1943:604
Bombus (Chromobombus) Dalla Torre, 1880:40, type-spe- cies Apis muscorum Linnaeus (= Bombus muscorum (Linnaeus)) by subsequent designation of Sandhouse, 1943:538
Bombus (Agrobombus) Vogt, 1911:52, type-species Apis agrorum Fabricius (=Bombus pascuorum (Scopoli)) by subsequent designation of Sandhouse, 1943:523
[Agrabombus Skorikov, 1914a:119, incorrect subsequent spelling]
Bombus (Ruderariobombus) Kriiger, 1920:350, type-spe- cies Apis ruderaria Miiller (= Bombus ruderarius (Miiller)) by subsequent designation of Yarrow, 197 1:27
Agrobombus (Adventoribombus) Skorikoy, 1922a:150, type-speciesA grabombus adventor Skorikov (=Bombus filchnerae Vogt) by subsequent designation of Sandhouse, 1943:522, new synonym
[Agrobombus (Adventoriobombus) Skorikoy, 1931:218, incorrect subsequent spelling]
Agribombus Skorikovy, 1938a:145, unjustified emendation
[Bombus (Thoraocbombus) Esmaili & Rastegar, 1974:52, incorrect subsequent spelling]
[Bombus (Thoracibombus) Schwarz et al., 1996:197, in- correct subsequent spelling]
@ TAXONOMIC STATUS. Richards (1968) treated Thoracobombus and Adventoribombus as separate subgenera, although he questioned whether they should be kept separate. I have followed Tkalcti (1974a) in treating B. adventor (=B. filchnerae) as part of a single subgenus Thoracobombus.
Bombus (Th.) filchnerae Vogt Filchnerae Vogt, 1908:100, examined adventor (Skorikov, 1914a:119 [Agrabombus}) lii Tkalcit, 1961b:355
Bombus (Th.) muscorum (Linnaeus)
Mujcorum (Linnaeus, 1758:579 [Apis]) examined
pallidus Evans, 1901:47, not of Cresson, 1863:92 (= B. pensylvanicus (DeGeer))
[fulvofasciatus Friese, 1905:520, infrasubspecific]
laevis Vogt, 1909:63
?nigripes Pérez, 1909:158, not of Haliday in Curtis ef al., 1837:321 (= B. dahlbomii Guérin-Méneville)
?pereziellus (Skorikoy, 1922a:150 [Agrobombus]) replace- ment name for nigripes Pérez, 1909:158
?bannitus (Skorikov in Popov, 1930:98 [Agrobombus])
?liepetterseni Loken, 1973:152
celticus Yarrow, 1978:15, replacement name for pallidus Evans, 1901:47
@ TAXONOMIC STATUS. B. bannitus (= B. smith- ianus of authors, a misidentification (=B. pascuorum)) has been regarded as a separate species by some authors (e.g. Richards, 1935; Tkalct, 1987; Rasmont & Adamski, 1995) on the basis of its semi-melanic colour pattern and more coarsely sculptured surface of
108
gastral terga [V-V. However, Loken (1973: fig. 81) found no difference between these taxa in a morphometric study (other authors reporting no clear morphological differences include Richards, 1935; Alford, 1975; Pekkarinen, 1979; Rasmont, 1982; Baker, 1996a). Furthermore, I have collected many specimens with a range of intermediate colour patterns on the Isle of Skye in western Scotland. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
B. pereziellus has also been regarded as a separate species by Rasmont & Adamski (1995), because of its dark colour pattern (even darker than B. bannitus, B. pereziellus has the thoracic dorsum black rather than red-brown, and has more black hairs on gastral tergum II, whereas these black hairs tend to be more frequent on tergum I for B. bannitus) and because it is endemic to the island of Corsica. Morphologically it was con- sidered by Rasmont (1982) to show no perceptible differences from B. muscorum or B. bannitus. Further- more, a male with a colour pattern apparently intermediate between B. muscorum and B. pereziellus is mentioned by Delmas (1976:271). Depending on the species concept embraced, some differences might be expected for a peripheral population such as this even if it were conspecific and I shall treat them as parts of a single variable species. Further evidence is awaited.
O NOMENCLATURE. Richards (1935, 1968), Yarrow (1968) and Léken (1973) recognised that none of the admissable syntypes in the Linnean collection agreed with the traditional interpretation of B. muscorum, which is very rare in the parts of Sweden where Linnaeus collected (Richards, 1935; Loken, 1973; Day, 1979), but took no action. When Day (1979) came to fix the application of the name, he had no reason to believe that Linnaeus had not described his A. muscorum from the syntype specimen that was subsequently described as lectotype (= B. humilis Illiger).
To reaffirm the traditional usage of B. muscorum, a case was made to ICZN by Loken et al. (1994). This sought an Opinion from ICZN (ICZN, 1996) that set aside by use of its Plenary Power (ICZN, 1985: Arti- cles 78b, 79) the lectotype designation forA. muscorum by Day from application of the Code (ICZN, 1985) and then designated a neotype (ICZN, 1996: 64) to conserve the traditional usage of the name for even the narrowest concept of the taxon (ICZN, 1985: Article 75).
Bombus (Th.) anachoreta (Skorikov) anachoreta (Skorikov, 1914a:121 [Agrobombus])
P.H. WILLIAMS
Bombus (Th.) opulentus Smith opulentus Smith, 1861:153, examined
Bombus (Th.) zonatus Smith zonatus Smith, 1854:389
Bombus (Th.) humilis MligerS fulvefcens (Schrank, 1802:367 [Apis]) humilis Wliger, 1806:171 ?tristis Seidl, 1837:69 ?variabilis Schmiedeknecht, 1878:424, not of Cresson, 1872:284 (= B. variabilis (Cresson)) ?subbaicalensis Vogt, 1911:42,54
OQ NOMENCLATURE. When Day (1979) came to fix the application of A. muscorum Linnaeus (see the comments on B. muscorum), he had no reason to believe that Linnaeus had not described this taxon from the syntype specimen that was subsequently described as lectotype (= B. humilis Mlliger). This action brought B. humilis Mliger into subjective junior synonymy with B. muscorum (Linnaeus).
To reaffirm the traditional usage of B. muscorum and B. humilis, a case was made to ICZN by Loken et al. (1994). This sought an Opinion from ICZN (ICZN, 1996) that set aside by use of its Plenary Power (ICZN, 1985: Articles 78b, 79) the lectotype designation forA. muscorum by Day from application of the Code (ICZN, 1985) and then designated a neotype (ICZN, 1996: 64) to conserve the traditional usage of B. muscorum and B. humilis ((CZN, 1985: Article 75).
However, Warncke (1986) recognised B. fulvescens (Schrank) as questionably conspecific with B. humilis. I have seen no type specimens, but the description is consistent with this interpretation. B. fulvescens is therefore likely to be the oldest available name for this species.
€ APPLICATION TO ICZN. Although B. fulvescens may be the oldest available name for the present interpretation of this species, the name B. humilis has been in common use for the species since 1947 (e.g. case and references in Lgken et al., 1994). In contrast, I know of no publications using the name B. fulvescens (Schrank) since 1947. Warncke (1986:98) followed the listing of this name with ‘Art. 23b’, which is a reference to purpose of the Principle of Priority (ICZN, 1985). I agree that, in the interests of stability, an application be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79) (see the comments on B. muscorum).
Bombus (Th.) deuteronymus Schulz senilis Smith, 1879:131, examined, not of Fabricius, 1775:382 (= B. pascuorum (Scopoli)) deuteronymus Schulz, 1906:267, replacement name for
CHECKLIST OF BUMBLE BEES
senilis Smith, 1879:131 velox (Skorikoy, 1914a:120 [Agrobombus}]) [superequester (Skorikov, 1914c:405 [Agrobombus]) infrasubspecific ] superequester (Skorikov, 1925:116 [Agrobombus}) bureschi Pittioni, 1939b:1, examined
Bombus (Th.) schrencki Morawitz Schrencki Morawitz, 1881:123 Schrencki Morawitz, 1881:250, redescribed konakovi Panfiloy, 1956:1330
?Bombus (Th.) honshuensis (Tkalcii) honshuensis (Tkalcti, 1968a:47 [Megabombus])
@ TAXONOMIC stTATUS. B. honshuensis and B. schrencki have allopatric distributions in northern Ja- pan (Sakagami & Ishikawa, 1969; Ito & Munakata, 1979: fig. 6; Ito, 1993), with B. honshuensis being possibly a disjunct peripheral population of B. schrencki. The two taxa are closely similar, and yet despite some variation in morphology, apparently con- sistent differences have been described (Tkalcii, 1968a; Sakagami & Ishikawa, 1972). Nonetheless, some dif- ferences might be expected even if they were conspecific, depending on the species concept ac- cepted (see the comments on B. ruderatus), so further evidence is awaited.
Bombus (Th.) impetuosus Smith impetuosus Smith, 1871:249, examined Potanini Morawitz, 1890:350, new synonym yuennanensis Bischoff, 1936:14, examined combai Tkalcti, 1961b:357, new synonym
@ TAXONOMIC STATUS. The white-banded B. potanini is morphologically closely similar to the yellow-banded B. impetuosus. Some individuals from Sichuan are intermediate in colour pattern in that they have the pale bands of the thorax and gastral tergum I white, and the pale band of tergum II yellow. There is considerable variation in the male gonostylus, but this variation appears to overlap between the the colour forms and I shall treat them as parts of a single variable species. S.-f. Wang and J. Yao (in litt.) also believe that the two taxa may be conspecific. Further evidence is awaited.
Bombus (Th.) remotus (Tkalcii) remotus (Tkalcti, 1968a:45 [Megabombus]) examined
Bombus (Th.) pseudobaicalensis Vogt Pseudobaicalensis Vogt, 1911:43,53 gilvus (Skorikov, 1925:117 [Agrobombus})
109 Bombus (Th.) hedini Bischoff
unicolor Friese, 1905:514, examined, not of Kriechbaumer, 1870:159 (= B. maxillosus Klug) hedini Bischoff, 1936:15
Bombus (Th.) ruderarius (Miiller) ruderaria (Miller, 1776:165 [Apis]) Derhamella (Kirby, 1802:363 [Apis]) examined montanus Lepeletier, 1836:463 simulatilis Radoszkowski, 1888:317, examined
Bombus (Th.) inexspectatus Tkalcit lutescens Kriiger, 1939:105, not of Pérez, 1890:154 (= B. flavidus Eversmann) inexspectatus (Tkalct, 1963:187 [Agrobombus}) [inexpectatus (Reinig, 1981:161 [Megabombus}]) incorrect subsequent spelling]
COMMENT. On the grounds of its peculiar morphol- ogy, B. inexspectatus has been suggested to be an obligate social parasite in colonies of other Bombus species, with B. ruderarius being the most likely host (Yarrow, 1970). As yet, there is no direct evidence for this behaviour (Rasmont, 1988). See the comments on the subgenus Psithyrus and on B. hyperboreus.
Bombus (Th.) sylvarum (Linnaeus) /ylvarum (Linnaeus, 1761:425 [Apis]) examined Daghestanicus Radoszkowski, 1877a:vii Dagestanicus Radoszkowski, 1877b:211, redescribed
Bombus (Th.) veteranus (Fabricius) veterana (Fabricius, 1793:324 [Apis]) arenicola Thomson, 1872:31
Bombus (Th.) mlokosievitzii Radoszkowski
Mlokosievitzii Radoszkowski, 1877a:viii
Mlokassewiczi Radoszkowski, 1877b:212, redescribed
pérezi Vogt, 1911:55, not of Schulthess-Rechberg, 1886:275 (= B. perezi (Schulthess-Rechberg))
vogtiellus (Tkalcu, 1977:224 [Megabombus]) replacement name for perezi Vogt, 1911:55
[mlokossowiczi (Reinig, 1981:161 [Megabombus]) incor- rect subsequent spelling]
O NOMENCLATURE. There are particularly many incorrect subsequent spellings of B. mlokosievitzii.
Bombus (Th.) pascuorum (Scopoli) Pafcuorum (Scopoli, 1763:306 [Apis]) Jenilis (Fabricius, 1775:382 [Apis]) agrorum (Fabricius, 1787:301 [Apis]) not of Schrank, 1781:397 (= B. mesomelas Gerstaecker) thoracicus Spinola, 1806:30
110
arcticus Dahlbom, 1832:50, not of Quenzel in Acerbi, 1802:253 (= B. hyperboreus Schonherr)
cognatus Stephens, 1846:17, examined
smithianus White, 1851:158
@ TAXONOMIC STATUS. Warncke (1986) listed B. cognatus as a synonym of B. muscorum, possibly following Stephens (1846), who wrote of B. cognatus: ‘Closely allied to Bo. Muscorum, of which the exam- ples I possess may be immature specimens’. Pagliano (1995) listed B. cognatus as a species separate from both B. muscorum and B. pascuorum, but without any explanation.
Saunders (1896:366—367) wrote ‘I have re-exam- ined the type of cognatus, Steph., . . . F. Smith placed it in the British Museum collection. ... Saunders considered this specimen to be conspecific with B. agrorum (Fabricius), continuing: ‘It is certainly not the species known on the Continent as cognatus’.
A female in the NHM collection bears the following labels: (1) a red-edged printed ‘Type’; (2) ‘cognatus.’ in handwriting identical to that of F. Smith; (3) ‘= agrorum / I.H.H.Y.’ in handwriting identical to that of I. Yarrow; (4) ‘B.M. Type / HYM. / 17B.1163’. I have examined this specimen and am unaware of any reason why it should not be considered the type of B. cogna- tus and (ignoring minor differences in colour pattern) conspecific with B. pascuorum.
OQ NOMENCLATURE. L@ken (1973) listed B. cogna- tus Stephens, 1846, as anomen nudum, citing Sherborn (1925). However, the reference by Sherborn is to an earlier paper by Stephens (1829), so this does not affect the use of the name B. cognatus Stephens, 1846.
Subgenus TRICORNIBOMBUS Skorikov Agrobombus (Tricornibombus ) Skorikoy, 1922a:151, type- species Bombus tricornis Radoszkowski by monotypy Bombus (Tricornibombus) Tkalcti, 1960:70
Bombus (Tr.) tricornis Radoszkowski tricornis Radoszkowski, 1888:319, examined
Bombus (Tr.) atripes Smith atripes Smith, 1852a:44, examined
P.H. WILLIAMS
Bombus (Tr.) imitator Pittioni imitator Pittioni, 1949:251, examined flavescens Pittioni, 1949:254, not of Smith, 1852a:45 (=B. flavescens Smith)
Subgenus FERVIDOBOMBUS Skorikov
Fervidobombus Skorikov, 1922a:153, type-species Apis fervida Fabricius (= Bombus fervidus (Fabricius)) by subsequent designation of Frison, 1927:69
Bombus (Fervidobombus) Franklin, 1954:47
Bombus (Digressobombus ) Laverty etal., 1984:1051, type- species Megabombus digressus Milliron (= Bombus digressus (Milliron)) by original designation
@ TAXONOMIC STATUS. The subgenus Digresso- bombus was described subsequent to Richards (1968). I have treated Digressobombus as part of a single subgenus Fervidobombus (Williams, 1995), as has Labougle (1990). While this study found no evidence for monophyly of the combined group, I believe that this is more likely (unpublished data) than monophyly of the subgenus Fervidobombus excluding Digresso- bombus.
COMMENT. This is the only early-diverging and large subgenus of bumble bees to occur in the New World other than the subgenus Psithyrus. Although it makes up only a small part of the fauna of America north of Mexico, it makes up most of the low- to mid-altitude bumblebee fauna of Central and SouthAmerica. It also includes the only species of bumble bees genuinely occurring in tropical lowland wet forest (e.g. Moure & Sakagami, 1962; Milliron, 1973a; Cameron & Whitfield, 1996). The species with more temperate distributions appear to occupy similar habitats and show similar flower-depth preferences to species of subgenera such as Thoracobombus and Megabombus in the Old World.
Bombus (Fy.) fervidus (Fabricius) feruida (Fabricius, 1798:274 [Apis]) ?Californicus Smith, 1854:400, examined Dumoucheli Radoszkowski, 1884:78 sonome Howard, 1902:pl. II
@ TAXONOMIC STATUS. B. fervidus and B.
CHECKLIST OF BUMBLE BEES
californicus have been regarded both as conspecific (e.g. Milliron, 1973a; Labougle, 1990) and as separate species (e.g. Franklin, 1913; Stephen, 1957; Thorp et al., 1983: Poole, 1996). Both Franklin (1913:239) and Stephen (1957) also considered the possibility that they are conspecific as quite reasonable.
Many specimens from the north west of North America show a reduction in the extent of the yellow bands on the scutellum and gastral terga I-III and appear to be intermediate or recombinant individuals. Indeed, Stephen’s (1957:32) figure 2 shows several patterns that could represent a continuum in variation between the two forms. Thorp ef al. (1983) found no intermediate females in California, although some of the males of B. californicus were said to approach the pattern of B. fervidus.
In view of the existence of apparent intermediates between these nominal taxa in at least part of their range, they are treated here as likely to be conspecific. More evidence is awaited.
OQ NOMENCLATURE. Apis feruida is the original spelling in Fabricius (1798). The orthography of this publication employs ‘u’ in place of “v’ widely, a com- mon practice of the period. This convention has since changed and subsequent authors have consistently used ‘vy’ for B. fervidus.
In fact, whatever the interpretation of the Code, pragmatically it matters little which spelling offervidus is used unless either of the spellings were to be pub- lished as the name of another taxon in Bombus. See the comments on the spelling of B. pensylvanicus.
Bombus (Fy.) pensylvanicus (DeGeer) penfylvanica (DeGeer, 1773:575 [Apis]) americanorum Fabricius, 1804:346 ?sonorus Say, 1837:413 pallidus Cresson, 1863:92 Pensylvanicus Cresson, 1863:94 flavodorsalis Franklin, 1913:409 pennsylvanicus Hurd, 1979:2204, unjustified emendation
@ TAXONOMIC sTATUS. B. pensylvanicus and B. sonorus have been regarded both as conspecific (e.g. Milliron, 1973a; Labougle ef al., 1985; Labougle, 1990; Poole, 1996) and as separate species (e.g. Franklin, 1913 [but see p. 239]; Stephen, 1957; Thorp et al., 1983; S. Cameron in litt.).
From the few males from the United States (not Mexico) that I have examined in detail, there appear to be subtle differences in the male genitalia (e.g. in the shape of the penis valve head). However, Labougle (1990) reports that the two ‘forms are geographically intermixed in México, and chromatically intermediate specimens occur, mainly in northeastern México and southwestern Texas’. He went on to say that ‘In fact, it is sometimes difficult to place a Mexican specimen in either subspecies because there are specimens with the
111
coloration of the scutellum and the punctation of the clypeus intermediate between the two taxa. Average differences of certain proportions are found . . . but do not differentiate all specimens’. G. Chavarria (pers. com.) also believes that intermediate specimens occur in Mexico and that they are conspecific. Taking an extreme viewpoint, it is even possible to see the ‘typi- cal’ B. sonorus colour pattern as intermediate between B. pensylvanicus (in the strict sense) and the extreme pale form that has the thoracic dorsum and gastral tergum I entirely yellow (flavodorsalis, see Thorp et al., 1983: fig. 137b).
In view of the existence of apparent intermediates between these nominal taxa in at least part of their range, they are treated here as likely to be conspecific. More evidence is awaited.
O NOMENCLATURE. Apis penfylvanica is the origi- nal spelling in DeGeer (1773). The orthography of this publication employs ‘Pin place of ‘s’ widely, a com- mon practice of the period. This convention has since changed and subsequent authors (e.g. Cresson, 1863) have consistently used ‘s’ for B. pensylvanicus.
Technically, according to the Code (ICZN, 1985: Article 32b), pensylvanicus with just two ‘n’s is the correct original spelling, to be preserved unaltered unless it is demonstrably incorrect under Article 32c. Article 32c(ii) states that clear evidence of an inadvert- ent error is only admissable if it lies within the original publication, without recourse to any external source of information (DeGeer, 1773, spelled Penjylvanie and penfylvanica consistently in this way). Any intentional change to that spelling in a subsequent publication is an unjustified emendation under Article 33b(iii).
In fact, whatever the interpretation of the Code, pragmatically it matters litthe which spelling of pensylvanicus is used unless either of the spellings were to be published as the name for another taxon in Bombus. No doubt many will prefer to use B. pennsylvanicus, although the name does appear as B. pensylvanicus in the recent checklist by Poole (1996) (and by analogy, the similar spelling of Vespula pensylvanica (Saussure) has been accepted, e.g. by Akre et al., 1980; Edwards, 1980).
COMMENT. This species was deliberately introduced into the Philippines, but is not known to have persisted (Frison, 1925b).
Bombus (Fy.) excellens Smith excellens Smith, 1879:133, examined
Bombus (Fy.) dahlbomii Guérin-Méneville Dahlbomii Guérin-Méneville, [1835, see Cowan, NO TE2 epi nigripes Haliday in Curtis et al., 1836:321
112
O NOMENCLATURE. Cowan (1971), considering Guérin-Méneville’s insect volume, states that ‘it is quite certain that valid publication [of the Insectes text] under the International Code of Nomenclature did not take place until August or September 1844.’ However, he lists plate 75, on which B. dahlbomii appears as figure 3 together with a legend containing the name, as having been published in livraison 39 in June 1835. This meets the criteria for valid publication (ICZN, 1985: Article 8). Therefore B. dahlbomii ts the oldest available name for this species.
Bombus (Fyv.) morio (Swederus)
morio (Swederus, 1787:283 [Apis]) examined
velutinus Mliger, 1806:175
violaceus Lepeletier, 1836:473
carbonarius Handlirsch, 1888:241, not of Menge, 1856:27 [fossil]
Kohli Cockerell, 1906:75, replacement name for carbonarius Handlirsch, 1888:241
Bombus (Fy.) diligens Smith diligens Smith, 1861:154, examined dolichocephalus Handlirsch, 1888:244
Bombus (Fy.) opifex Smith opifex Smith, 1879:133, examined
Bombus (Fy.) rubriventris Lepeletier rubriventris Lepeletier, 1836:472, examined
@ TAXONOMIC sTATUS. B. rubriventris is known from a single female specimen from ‘St. Domingue’ (2= Sao Domingos, Goids) (Milliron, 1973a). This specimen has dark brown wings and the pubescence is black, except that most of the hairs of the thorax are grey-tipped, and the hairs of gastral terga II-IV are bright ‘coppery’ red.
This colour pattern resembles the Andean B. excellens, although the pubescence of B. rubriventris is much shorter and more even; the oculo-malar area is nearly square rather than nearly twice as long as the basal breadth of mandible; and tergum VI is raised subapically. Franklin (1913) had not seen B. rubriventris but suggested that it was probably a ‘freak specimen’ of B. carolinus (a misidentification, = B. excellens). Milliron (1973a) had examined B. rubriventris and considered the morphological char- acters to be very much like those of B. bellicosus. However, B. rubriventris can be distinguished by the much finer punctures in the centre of the clypeus and by an absence of a median ridge on tergum VI. I consider B. rubriventris to be more similar in these characters to B. opifex, although it can be distin-
P.H. WILLIAMS
guished from that species by a pair of characteristi- cally slightly recessed bands of fine punctures extending anteriorly from the ocello-ocular areas and by a shallow median groove in the subapically raised area of tergum VI.
The colour pattern is very distinctive among non- Andean bumble bees in South America and does not appear to be the result of abnormal colour develop- ment. The specimen has had the gaster glued back into place, although the characters of both the head and gaster appear to be distinctive, so there is no reason to believe that the specimen is a composite and not genuine.
COMMENT. Milliron (1973a) researched the history of this specimen and believed that it may have been collected as early as 1800. He concluded that it was probably a highland species from southeastern Brazil and that it may now be extinct. If so, and accepting that it is very difficult to establish the absence of a species, this would be one of the few known cases of complete extinction of an insect species.
Bombus (Fyv.) bellicosus Smith thoracicus Sichel, 1862:121, not of Spinola, 1806:30 (=B. pascuorum (Scopoli)) bellicosus Smith, 1879:131, examined Emiliae Dalla Torre, 1890:139, replacement name for thoracicus Sichel, 1862:121
Bombus (Fy.) pullatus Franklin pullatus Franklin, 1913:122
Bombus (Fy.) transversalis (Olivier) tran/verfalis (Olivier, 1789:65 [Apis]) Cajennenjis (Fabricius, 1798:273 [Apis]) incarum Franklin, 1913:131
Bombus (Fy.) atratus Franklin® azurea (Christ, 1791:129 [Apis]) ?atratus Franklin, 1913:118, not of Friese, 1911:572 (=B. mucidus Gerstaecker) (provisional synonym) ?niger Franklin, 1913:120, examined (provisional syno- nym) ?nigriventris Friese, 1913:87 (provisional synonym)
@ TAXONOMIC stTaTUS. Atleast four species of the subgenus Fervidobombus from Central and South America have many individuals for which the pubes- cence is almost entirely black. The genitalia of the males are quite distinctive, but association of the conspecific females with these males has caused prob- lems.
In the original description of B. niger, Franklin stated that ‘atratus is possibly the male of niger (p.
CHECKLIST OF BUMBLE BEES
121), whereas in the original description of B. atratus he stated both that ‘Niger may represent the females of this species’ (p. 118) and that “This may be the true male of kohli’ (p. 119). B. niger was described from a syntype series of four queens and four workers, of which one queen in the Smithsonian collection carries, amongst others, a red label ‘LECTOTYPE / Bombus / niger Franklin/H.E.Milliron ‘59’ and a label “Boquete / Chiriqui’. This Central American locality was men- tioned by Franklin, but is outside the known distribution of the species (Milliron, 1973a) to which the specimen belongs. In my opinion, this lectotype of B. niger is not conspecific with B. pullatus (contrary to the sugges- tion by Labougle, 1990, see also Milliron, 1962) but rather is conspecific with B. atratus Franklin.
Another possibility is that this variable species is the Apis azurea of Christ (1791). I know of no type specimens and the type locality was said to be inAfrica (Ist in Afrika am Vorgebiirg der guten Hofnung zu Haus’). The description and figure of the colour pat- tern do not agree with any African bees that Ihave been able to trace, but do resemble closely the yellow- banded individuals of the SouthAmerican B. niger, the Central American B. medius Cresson, and the South American B. transversalis (Olivier) (although for the last named species the yellow bands on the thorax are usually broader). Among the specimens to hand, the wings do appear slightly more “Schwarzblaue’ for B. niger, as described for A. azurea, although these grounds seem slim justification from which to estab- lish the application of a name.
O NOMENCLATURE. B. azureus is possibly the old- est available name for this species.
Milliron (1962), without mention of the name B. azureus, first regarded B. atratus and B. niger as conspecific and, following the Principle of First Re- viser (ICZN, 1985: Article 24), chose B. atratus as the valid name for the species.
Unfortunately, B. atratus Franklin, 1913, is a junior primary homonym of B. mucidus var. atratus Friese, 1911 (deemed to be subspecific, see ICZN, 1985: Article 45g(ii)), therefore the name B. atratus Franklin is invalid (ICZN, 1985: Article 57b).
© APPLICATION TOICZN. The name B. azureus has not been used since the original publication. The name B. atratus has been used for this species since 1947 (e.g. Moure & Sakagami, 1962; Sakagami & Zucchi, 1965; Sakagami ef al., 1967; Milliron, 1971, 1973a; Sakagami, 1976; Ito, 1985; Labougle, 1990; Varela, 1992; Silveira & Cure, 1993). It is suggested that, in the interests of stability (ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to suppress both the unused senior synonym (ICZN, 1985: Article 79) and the senior homonym. This would achieve both an unambiguous, valid name for this species (see the comments on B. muscorum) and also
113
help to protect the validity of the names B. medius and B. transversalis from future change. However, the consequence of this action would be that atratus Friese would no longer be available for a subspecies of B. mucidus.
Bombus (Fv.) digressus (Milliron) digressus (Milliron, 1962:730 [Megabombus]) examined
Bombus (Fy.) brasiliensis Lepeletier brasiliensis Lepeletier, 1836:470, examined
Bombus (Fy.) steindachneri Handlirsch Steindachneri Handlirsch, 1888:239
@ TAXONOMIC sTATUS. B. medius andB. steindac- hneri have been regarded both as separate species (Milliron, 1973a; Labougle, 1990) and as conspecific (G. Chavarria, pers. com.).
Labougle (1990) reports that ‘Although the chro- matic differences between B. medius and B. steindachneri are conspicuous, the male genitalia are extremely similar’. Labougle listed four character dif- ferences from the male genitalia and I can confirm two of these: (1) that the head of the penis valve of B. steindachneri has fewer fine teeth or serrations; and (2) that the interior process of the volsella (misinter- preted as the preapical tooth of the “gonostylus’; for discussion of homologies see Williams, 1991) of B. steindacheri is narrower. However, I have examined only a few males and these characters might be ex- pected to vary among other individuals. Labougle (1990) continued: ‘The lack of chromatic and morpho- logical intermediates supports the idea of two different species’.
Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as two separate species.
Bombus (Fy.) medius Cresson medius Cresson, 1863:97
Bombus (Fy.) weisi Friese laboriosus Smith, 1861:153, examined, not of Fabricius, 1804:352 (= Emphoropsis laboriosus (Fabricius)) weisi Friese, 1903:253, examined nigrodorsalis Franklin, 1907:90
O NOMENCLATURE. The lectotype female of B. weisi by designation of Milliron (1960:98) was recog- nised as conspecific with B. nigrodorsalis by Labougle (1990) (Ihave examined the lectotype of B. weisi at the MNHU, Berlin, and agree with Labougle). He then
114
used B. weisi (the oldest available name) as the valid name for this species. However, a case could be made in favour of the use of either name.
For Labougle’s (1990) use of this previously unused senior synonym to be considered by ICZN as a prima facie case of upsetting the use of a long-accepted name in its accustomed meaning (ICZN, 1985: Article 23b), the name B. weisi should not have been used in this sense in the preceding fifty years; and at least five authors should have used the junior name, B. nigrodorsalis, in at least ten publications during the same period (ICZN, 1985: Article 79c). As far as 1am aware, no other admissable publications have used B. weisi (Williams, 1995, disclaimed any nomenclatural action in a list of names for material examined), al- though publications using the junior name B. nigrodorsalis Franklin for this species since 1947 are more common, including Milliron (1961, 1962, 1971, 1973a), Laverty et al. (1984), Labougle et al. (1985), Williams (19855) andAsperen de Boer (1992b). Other such references may exist, therefore this may be seen as a borderline case, requiring an application to be made to ICZN to use its Plenary Power if suppression of the unused senior synonym, B. weisi, is required (see the comments on B. muscorum).
On the other hand, a change of valid name from B. nigrodorsalis to B. weisi does not appear to be a serious disruption of common usage, so there is no obvious need for action to retain B. nigrodorsalis and I have continued to use B. weisi.
Bombus (Fy.) trinominatus Dalla Torre modestus Smith, 1861:153, examined, not of Eversmann, 1852:134 (= B. modestus Eversmann) trinominatus DallaTorre, 1890:139, replacement name for modestus Smith, 1861:153 xelajuensis Asperen de Boer, 1992b:162, examined (pro- visional synonym)
@ TAXONOMIC STATUS. The description of B. xelajuensis shows that this nominal taxon, known from a single location, diverges only slightly in col- our pattern and morphology from the otherwise restricted and uncommon mountain species B. trinominatus. Therefore it seems most likely to be conspecific with B. trinominatus, with a slightly dif- ferent colour pattern. However, the information available at present is not conclusive, and it remains possible that it represents a separate species, and further evidence is awaited.
Bombus (Fy.) mexicanus Cresson mexicanus Cresson, 1878:187
P.H. WILLIAMS
Bombus (Fyv.) brevivillus Franklin brevivillus Franklin, 1913:119 ?abditus (Tkalcit, 1966:271 [Megabombus}))
@ TAXONOMIC STATUS. The single known female of B. abditus was described as originating from “Rep. de Guinée Beyla’ (equatorial Africa). However, it is indistinguishable from B. brevivillus according to Sakagami (1976:427) and probably represents an in- troduced or mislabelled individual (Michener, 1979).
Subgenus SENEXIBOMBUS Frison Bremus (Senexibombus) Frison, 1930:3, type-species Bombus senex Vollenhoven by original designation [Bombus (Senecibombus) Kruseman, 1952:101 incorrect subsequent spelling] Bombus (Senexibombus) Richards, 1968:217
Bombus (Sx.) kulingensis Cockerell kulingensis Cockerell, 1917:266 tajushanensis Pittion1, 1949:244
Bombus (Sx.) bicoloratus Smith bicoloratus Smith, 1879:132, examined
Bombus (Sx.) senex Vollenhoven Senex Vollenhoven, 1873:229
Bombus (Sx.) irisanensis Cockerell irisanensis Cockerell, 1910a:416, examined
CHECKLIST OF BUMBLE BEES
Subgenus DIVERSOBOMBUS Skorikov Bombus (Diversobombus ) Skorikov, 1914c:406, type-spe- cies Bombus diversus Smith by subsequent designation of Sandhouse, 1943:546 Diversibombus Skorikov, 1938b:2, unjustified emenda- tion
Bombus (Dy.) trifasciatus Smith trifasciatus Smith, 1852a:43, examined montivagus Smith, 1878:168, examined montivagus Smith, 1879:131, redescribed ?wilemani Cockerell, 1911:100, examined albopleuralis Friese, 1916:108, examined ?maxwelli Pendlebury, 1923:67, examined mimeticus Richards, 1931b:529, examined malaisei (Skorikov, 1938b:2 [Diversibombus]) not of Bischoff, 1930:4 (= B. sporadicus Nylander) atropygus (Tkalcti, 1989:58 |Megabombus]}) examined
@ TAXONOMIC sTaTUS. Several of these nominal taxa have been treated as separate species, for example as B. albopleuralis (= B. mimeticus) {Himalaya}, B. montivagus {northern Burma to southern China], B. maxwelli [Peninsular Malaysia] and B. wilemani [Tai- wan] (Tkalcii, 1968b, 1989). However, aside from differences in colour pattern (Fig. 13), they are closely similar in morphology and show a range of variation (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of varia- tion, I shall continue to treat them as parts of a single variable species.
Bombus (Dyv.) longipes Friese longipes Friese, 1905:511 hummeli Bischoff, 1936:18, examined
Bombus (Dy.) diversus Smith diversus Smith, 1869:207, examined tersatus Smith, 1869:207, examined
OQ NOMENCLATURE. Tkalcti (1965) first explicitly regarded B. diversus and B. tersatus as conspecific and, following the Principle of First Reviser (ICZN, 1985: Article 24), chose B. diversus as the valid name for the species.
115
Bombus (Dv.) ussurensis Radoszkowski Ussurensis Radoszkowski, 1877b:196 {ussuriensis Morawitz, 1881:254, incorrect subsequent spelling]
Bombus (Hortobombus) Vogt, 1911:56, type-species Apis hortorum Linnaeus (= Bombus hortorum (Linnaeus)) by subsequent designation of Sandhouse, 1943:559
Bombus (Odontobombus) Kriiger, 1917:61,65 (proposed as a section name but stated by Milliron, 1961:53, to be equivalent to his concept of the subgenus Megabombus Dalla Torre), type-species Apis argillacea Scopoli (= Bombus argillaceus (Scopoli)) by subsequent designa- tion of Williams, 1995:339
[Nortobombus Skorikoy, 1922b:map 3, incorrect subse- quent spelling]
Hortibombus Skorikov, 1938a:146, unjustified emenda- tion
Bombus (Mg.) supremus Morawitz supremus Morawitz, 1886:196 linguarius Morawitz, 1890:351
Bombus (Mg.) gerstaeckeri Morawitz Gerstdckeri Morawitz, 1881:242 Gerstaeckeri Hoffer, 1883:55, mandatory correction (ICZN, 1985: Article 32d)
Bombus (Mg.) consobrinus Dahlbom consobrinus Dahlbom, 1832:49
Bombus (Mg.) tichenkoi (Skorikoy) [tichenkoi (Skorikoy, 1922a:156 [Hortobombus]) published without description] tichenkoi (Skorikov, 1925:115 [Hortobombus])
116
?yezoensis Matsumura, 1932:pl. 1 ?przewalskiellus (Skorikov, 1933a:59 [Hortobombus]) kurilensis Sakagami, 1954:92
@ TAXONOMIC STATUS. B. tichenkoi and B. yezoensis have apparently been regarded both as conspecific and as separate species.
Sakagami (1954) described kurilensis as a (semi- melanic) subspecies of what he called ‘B. tersatus’ (a misidentification, = B. diversus). Following Tkalcit (1962), Sakagami subsequently identified his ‘B. tersatus’ as B. yezoensis (Ito & Sakagami, 1980). In this later paper he went on to recognise kurilensis as conspecific with B. tichenkoi, but then, despite having regarded kurilensis as a subspecies of what he now believed to be B. yezoensis in the earlier paper, listed B. tichenkoi as a species separate from B. yezoensis. The apparent contradiction was not explained, al- though morphological comparisons between these taxa, B. argillaceus and B. sushkini were tabulated (which show primarily that B. argillaceus 1s very different). He even noted the allopatric distributions of B. tichenkoi and B. yezoensis between the northern and southern Kurile Islands and the ‘resemblance of tichenkoi and dark individuals of yezoensis, especially in workers.’
Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat B. tichenkoi and B. yezoensis as parts of a single variable species.
Bombus (Mg.) sushkini (Skorikov) [saltuarius (Skorikov, 1922a:156 [Hortobombus]) pub- lished without description] sushkini (Skorikoy, 1931:235 [Hortobombus]) examined saltuarius (Skorikov, 1931:235 |Hortobombus])
@ TAXONOMIC STATUS. B. sushkini and B. saltuarius have been regarded both as conspecific (Bischoff, 1936) and as separate species (Skorikov, 1931; Tkalct, 1974a). I have as yet seen no evidence that more than one species is involved. More evidence is awaited.
O NOMENCLATURE. Skorikov (1931) provided the first valid publication of the names B. sushkini and B. saltuarius as two separate species. Subsequently, Bischoff (1936) regarded the two as conspecific and, following the Principle of First Reviser (ICZN, 1985: Article 24), chose B. sushkini as the valid name for the species.
Bombus (Mg.) portchinsky Radoszkowski Portchinskij Radoszkowski, 1883:208 Portchinsky Radoszkowski, 1883:208[210], 209[211], 210[212]
O NOMENCLATURE. Radoszkowski (1883) pub- lished two different spellings of B. portchinsky,
P.H. WILLIAMS
repeating this second form several times (page num- bers 207 and 208 are repeated twice for different pages, so the two spellings do not occur on the same page). Precedence of the correct original spelling should be determined using the Principle of the First Reviser (ICZN, 1985: Article 24), but to date I have not found an author who has correctly cited both names and then chosen one in precedence to the other (there are many incorrect subsequent spellings, e.g. Dalla Torre, 1896). Since the second form of the name has been used more recently (e.g. Baker, 1996b), I suggest that it should be given precedence. This form is a simple noun in apposition and so retains the same ending whatever the gender of the generic name with which it is combined (ICZN, 1985: Article 31b(11)).
Bombus (Mg.) hortorum (Linnaeus)
hortorum (Linnaeus, 1761:424 [Apis]) examined
meridionalis Dalla Torre, 1879:13
hispanicus Pittioni, 1939c:244, not of Friese, 1911:571 (= B. monticola Smith)
asturiensis (Tkalct, 1975:181 [Megabombus]) replace- ment name for hispanicus Pittioni, 1939c:244
?reinigiellus (Rasmont, 1983:43 [Megabombus])
@ TAXONOMIC STATUS. The SpanishB. asturiensis has been considered a separate species from B. hortorum by Rasmont (1983, 1988), although they have been treated as conspecific by Pittioni (1939c), Tkalcti (1975), Ornosa (1986a, 1986b, 1991), Castro (1988, 1993) and, more recently, by Rasmont ef al. (1995). The two taxa are closely similar.
The Spanish B. reinigiellus has also been consid- ered both as conspecific with B. hortorum (Castro, 1987) and as a separate species (e.g. Rasmont, 1983; Castro, 1988; Ornosa, 1991).The two taxa are allopatric (Rasmont, 1983), with B. reinigiellus being narrowly restricted to the Sierra Nevada of Spain, possibly as a disjunct peripheral population. B. reinigiellus is closely similar to B. hortorum, although subtle differences in characters of colour and morphology have been de- scribed (e.g. Rasmont, 1983; Castro, 1988; Ornosa, 1991). From the material I have examined, the mor- phological differences appear to be analogous to the variation between mainland and island populations of B. terrestris (see the comments on B. terrestris).
Depending upon the species concept embraced, such subtle differences as those between B. reinigiellus and B. hortorum might be expected even within a single population and [| shall treat all three taxa as conspecific for the present. More evidence is awaited.
COMMENT. B. hortorum has been introduced into New Zealand (e.g. Gurr, 1957; Macfarlane & Gurr, 1995) (see the comments on B. ruderatus, B. subterraneus and B. terrestris). It occurs in Iceland, where it has also probably been introduced (Prys- Jones et al., 1981) (see the comments on B. Jucorum).
CHECKLIST OF BUMBLE BEES
Bombus (Mg.) argillaceus (Scopoli) Argillacea (Scopoli, 1763:305 [Apis]) ligusticus Spinola, 1806:29
?Bombus (Mg.) ruderatus (Fabricius) ruderata (Fabricius, 1775:380 [Apis]) examined Perniger (Harris, 1776:131 [Apis]) villarricaensis Asperen de Boer, 1992a:133
@ TAXONOMICSTATUS. B. argillaceus and B. rude- ratus are similar in most characters and differ principally in the colour patterns of the queens (e.g. Reinig, 1939; Fig. 11). Scholl, Obrecht & Zimmer- mann (1992) found that hybrid queens between B. argillaceus and B. ruderatus do occur in parts of southeastern France, but are very rare. Whether or not the taxa on either side of this hybrid zone are consid- ered to be separate species therefore depends on which species concept is preferred. Because Scholl, Obrecht & Zimmermann (1992) estimated that only slight gene flow is occurring, I shall continue to treat them as separate species.
COMMENT. This species has been introduced into New Zealand (e.g. Gurr, 1957; Macfarlane & Gurr, 1995) (see the comments on B. hortorum, B. subterraneus and B. terrestris) and Chile (Arretz & Macfarlane, 1982; Asperen de Boer, 1993b). B. ruderatus also occurs on the Azores (which have never had a continental connection), where it may be pre- sumed to be an introduction (Yarrow, 1967).
Bombus (Mg.) czerskii Skorikov czerskii Skorikov, 1910b:413, examined
Bombus (Mg.) koreanus (Skorikov) koreanus (Skorikoy, 1933a:59 [Hortobombus]) pekingensis Bischoff, 1936:21, examined ?notocastaneus Tkalcit, 1961a:52 (provisional synonym)
@ TAXONOMIC STATUS. B. notocastaneus was de- scribed from a single male from Hubei. From the description, it appears most likely to be conspecific with B. koreanus.
Bombus (Mg.) melanopoda Cockerell
melanopoda Cockerell, 1910a:416, examined
@ TAXONOMIC STATUS. B. melanopoda is known from a single female specimen (labelled ‘Sumatra’, reverse *92.182.’) in the NHM collection in London.
According to the accessions catalogue, the number on the label of this specimen refers to 8 Hymenoptera presented in November 1892 by H. O. Forbes. The area of origin is given as Borneo, although this has
117
been crossed out and Sumatra added. Forbes’ (1885) account of his travels of 1878-1883 in Indonesia shows that he did not visit Borneo, although he did visit the mountains of southern Sumatra. There is no direct account of the collection of this specimen, although he recorded bees from at least three possible localities at higher altitudes: first, in late 1880 he climbed Gunung Tenggamus, where he recorded (p. 159) ‘a few bees’ at 7200 ft (2160 m); second, in 1881 he visited Gunung Dempa, where he recorded (p. 208) ‘a fine grey-haired humble-bee (Bombus senex)’ (iden- tification by Forbes) between 7000-7700 ft (2100-2310 m); and later in the same year, near the summit of Gunung Kaba (1983 m), he recorded again (p. 228) ‘A large humble-bee (Bombus senex)’.
B. melanopoda appears to be a morphologically distinct species. It can be distingished from the only other long-tongued bumblebee species known from Sumatra, B. senex, by the longer oculo-malar area of B. melanopoda, which is more than 1.5 times longer than the basal breadth of the mandible. Otherwise, the most closely related long-tongued bumble bee in any neighbouring area is B. trifasciatus from the Cameron Highlands of Peninsular Malaysia. How- ever, like most queens of the subgenus Megabombus, the holotype of B. melanopoda is easily distin- guished by its narrow longitudinal median groove subapically on gastral tergum VI (for B. trifasciatus this area is uniformly convex). The colour pattern of the B. melanopoda female is predominantly black, but the hairs of terga [V—V are very slightly paler, so there is some similarity to the darkest queens of B. koreanus, which have this pubescence brownish cream. However, unlike the few queens of B. koreanus available to me, the type of B. melanopoda has the unpunctured areas around the ocelli extend- ing to less than half the ocello-ocular distance; the dorsal furrow of the gena (between the vertex and the post-ocular area) is strongly marked anteriorly; and the dorsal face of the labral tubercles (the face adjacent to the clypeus) is more sharply separated from the anterior ventral face and more densely marked by moderate-sized punctures.
Since no further individuals have been found, the possibility that the holotype of B. melanopoda is a mislabelled melanic specimen of another species of the subgenus Megabombus ought to be explored, per- haps initially through a morphometric analysis.
COMMENT. Like most other Sumatran bumble bees (with the notable exception of the extensively greyish- white queens of B. senex, see Sianturi et al., 1995), the holotype of B. melanopoda is almost entirely black. It is likely that females of this species would be particu- larly easily mistaken for black individuals of B. senex (although males of B. melanopoda might be paler, as for B. koreanus).
118
Bombus (Mg.) securus (Frison) securus (Frison, 1935:346 [Bremus]) examined yuennanicus Bischoff, 1936:23, examined
Subgenus RHODOBOMBUS Dalla Torre
Bombus (Rhodobombus) Dalla Torre, 1880:40, type-spe- cies Bremus pomorum Panzer (= Bombus pomorum (Panzer)) by subsequent designation of Sandhouse, 1943:596
Bombus (Pomobombus) Kriiger, 1917:65, type-species Bremus pomorum Panzer (= Bombus pomorum (Pan- zer)) by subsequent designation of Sandhouse, 1943:589
Pomibombus Skorikov, 1938a:145, unjustified emenda- tion
Bombus (Rh.) armeniacus Radoszkowski armeniacus Radoszkowski, 1877b:202
Bombus (Rh.) mesomelas Gerstaecker® Agrorum (Schrank, 1781:397 [Apis]) arvenjfis (Gmelin in Linnaeus, 1790:2786 [Apis]) unjusti- fied replacement name for agrorum Schrank, 1781:397 mesomelas Gerstaecker, 1869:321
O NOMENCLATURE. Warncke (1986) listed B. agrorum (Schrank) as questionably conspecific with B. distinguendus, but without any explanation. Al- though I know of no extant type specimens, Schrank’s (1781) description of his B. agrorum of ‘Habitat rurv from Austria appears to me to be almost certainly of the same species as B. mesomelas, because the head is described as black and the pale hairs of the thorax and of gastral tergum I are described as grey, with the remainder of the gaster rusty or tawny-yellow (the head and the pale pubescence of the thorax and gaster are more uniformly dull yellowish fa B. distinguendus). See the comments onB. distinguendus.
& APPLICATION TO ICZN. Although B. agrorum is the oldest available name for the present interpretation of this species, the name B. mesomelas has been in
P.H. WILLIAMS
common use for the species since 1947 (e.g. Tkalcii, 1969, 1975; Delmas, 1976; Reinig, 1974, 1981; Ozbek, 1983; Rasmont, 1983; Ornosa, 1986a, b; Rasmont et al., 1987, 1995). I know of no publications using the name B. agrorum (Schrank) since 1947, although the name B. agrorum (Fabricius) was in widespread use for another species (= B. pascuorum) until Richards (1968). It is suggested that, in the interests of stability (ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to suppress the unused senior synonym, B. agrorum, and its unjustified re- placement name, B. arvensis (ICZN, 1985: Article 79) (see the comments on B. muscorum).
Bombus (Rh.) pomorum (Panzer) pomorum (Panzer, 1805(86):18 [Bremus]) Lefebvrei Lepeletier, 1836:461
Subgenus KALLOBOMBUS Dalla Torre
Bombus (Kallobombus) Dalla Torre, 1880:40, type-spe- cies Apis soroeensis Fabricius (= Bombus soroeensis (Fabricius)) by subsequent designation of Sandhouse, 1943:561
Bombus (Callobombus) Dalla Torre, 1896:503, unjustified emendation
Bombus (Soroeensibombus) Vogt, 1911:63, type-species Apis soroeensis Fabricius (= Bombus soroeensis (Fabricius)) by monotypy
[Bombus (Soroensibombus) Ball, 1914:78, incorrect sub- sequent spelling]
[Sorocoénsibombus Skorikoy, 1922a: map 15, incorrect subsequent spelling]
Bombus (KI.) soroeensis (Fabricius) @
Cardui (Miller, 1776:165 [Apis])
Joroeenfis (Fabricius, [1777, see Baker, 1996a:9]:246 [Apis})
Proteus Gerstaecker, 1869:325
perplexus Radoszkowski, 1884:82, not of Cresson, 1863:91 (= B. perplexus Cresson)
Radoszkowskyi Dalla Torre, 1890:139, replacement name for perplexus Radoszkowski, 1884:82
miniatocaudatus Vogt, 1909:56
© NOMENCLATURE. Baker(1996a) has established
CHECKLIST OF BUMBLE BEES
that the name B. cardui has narrow priority over B. soroeensis by publication date.
© APPLICATION TOICZN. AlthoughB. cardui is the oldest available name for the present interpretation of this species, the name B. soroeensis has been in com- mon use for the species since 1947 (e.g. Tkalct, 1969, 1975; Lgken, 1973; Alford, 1975; Delmas, 1976; Pekkarinen, 1979; Reinig, 1981; Ozbek, 1983; Rasmont, 1983; Ito, 1985; Ornosa, 1986a; Rasmont et al., 1995). The only publications using the name B. cardui since 1947 are those of Baker (1996a, b). Using this name contrary to the purpose of priority is not accepted as usage in the sense of the Code (ICZN, 1985: Article 23b), and so cannot justify the continued use of the name B. cardui in place of B. soroeensis. It is suggested that, in the interests of stability, an appli- cation be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79) (see the comments on B. muscorum).
COMMENT. Species of the subgenus Alpinobombus make up the most northerly distributed of all bee faunas (e.g. K. W. Richards, 1973). Indeed, three of the five species have a nearly circumpolar distribution, as a major component of an Arctic bumble bee fauna (Williams, 1996b). This relative homogeneity of the Arctic fauna among northern continents resembles the pattern in the Arctic flora, which shows little regional differentiation in comparison with more southern flo- ras (Hooker, 1861; Walker, 1995).
Bombus (Al.) hyperboreus Schonherr Arctica (Quenzel in Acerbi, 1802:253 [Apis]) hyperboreus Schénherr, 1809:57, unjustified replacement name for arcticus Quenzel, 1802:253 clydensis Yarrow, 1955:151, examined
119
@ TAXONOMIC STATUS. The identity of B. arcticus (Quenzel) has been uncertain. Warncke (1986) listed B. arcticus (Quenzel) as conspecific withB. lapponicus without any explanation. Presumably this was because B. lapponicus is extensively pale on the dorsum, al- though the pale pubescence is differentiated into yellow and red areas and much of it is much paler than Quenzel’s description. | agree with Lgken (1973) that, from the original description and the illustration (no type specimen is known to exist), B. arcticus (Quenzel) is most likely to be conspecific with B. hyperboreus, which has the pale pubescence uniformly brownish yellow.
OQ NOMENCLATURE. The name B. arcticus has rarely been used for this species in preference to B. hyperboreus, and perhaps only as a misidentification of B. arcticus Kirby (see e.g. Franklin, 1913; Richards, 193 1a). Loken (1973) considered B. arcticus (Quenzel) to be a nomen oblitum, so she continued to use the name B. hyperboreus. However, nomina oblita are not supported for a publication of this date by the present Code (ICZN, 1985: Article 79c(ii1)), although it does allow that B. arcticus (Quenzel) could be suppressed by use of the Plenary Power. See the comments on B. polaris Curtis.
€ APPLICATION TO ICZN. Although B. arcticus is the oldest available name for the present interpretation of this species, the name B. hyperboreus has been in common use for the species since 1947 (e.g. Loken, 1973; Milliron, 1973a; K.W. Richards, 1973; Svensson & Lundberg, 1977; Hurd, 1979; Pekkarinen, 1979; Pekkarinen ef al., 1981; Reinig, 1981; Rasmont, 1983; Pekkarinen & Teras, 1993). It is suggested that, in the interests of stability (ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79), in order to confirm usage of B. hyperboreus as the valid name (see the comments on B. muscorum).
COMMENT. B. hyperboreus has been suggested to be a social parasite in colonies of B. polaris, at least facultatively in some parts of its range (Milliron & Oliver, 1966; Loken, 1973: K. W. Richards, 1973). See the comments on Psithyrus and B. inexspectatus.
Bombus (Al.) balteatus Dahlbomes balteatus Dahlbom, 1832:36 nivalis Dahlbom, 1832:40 tricolor Dahlbom, 1832:41 ?Kirbiellus Curtis in Ross, 1835:1xii kirbyellus Dalla Torre, 1896:527, unjustified emendation tristis Sparre-Schneider in Friese, 1902:495, not of Seidl, 1837:69 (= B. humilis Iliger)
@ TAXONOMIC status. B. balteatus and B. kirbi- ellus have been considered conspecific by most authors (e.g. Thomson, 1872; Richards, 1931a; Skorikov, 1937;
120
Pittioni, 1942; Loken, 1973; Hurd, 1979; Thorp et al., 1983), although Milliron (1973a) considered them to be separate species that co-occur in some areas, par- ticularly in Alaska.
Milliron (1973a) described several characters by which to discriminate B. balteatus and B. kirbiellus, placing particular emphasis on the shape of male gastral sternum VIII and the female malar area.
From the small samples I have examined, I have been unable to find convincing evidence of discrete differences in these characters. Until more evidence to the contrary is available from critical studies of pat- terns of variation, I shall treat them as parts of a single variable species.
O NOMENCLATURE. Richards (193 1a) believed B. balteatus, B. nivalis and B. tricolor to be conspecific and selected the name B. balteatus to have precedence because it was published on an earlier page (page priority is not a mandatory part of the Code, only a recommendation, see ICZN, 1985: Recommendation 24A). However, Thomson (1872:35) had already cho- sen the name B. nivalis in precedence to B. balteatus and, following the Principle of the First Reviser (ICZN, 1985: Article 24), Thomson’s action should now stand. Consequently, the valid name for this species is B. nivalis, although the Code (ICZN, 1985) allows that this name could be suppressed by use of the Plenary Power.
€ APPLICATIONTOICZN. AlthoughB. nivalis is the valid name for this species, the name B. balteatus has been in common use for the species since 1947 (e.g. Loken, 1973; Milliron, 1973a; Plowright & Stephen, 1973; Hurd, 1979; Pekkarinen, 1979; Reinig, 1981; Rasmont, 1983; Thorp er al., 1983; Laverty & Harder, 1988; Pekkarinen & Teras, 1993). Itis suggested that, in the interests of stability, an application be made to ICZN to use its Plenary Power to suppress the unused name (ICZN, 1985: Article 79) (see the comments on B. muscorum). However, the consequence of this action would be that nivalis Dahlbom would no longer be available for a subspecies of B. balteatus.
Bombus (Al.) neoboreus Sladen strenuus Cresson, 1863:102, not of Harris, 1776:131 (=B. lapidarius (Linnaeus)) neoboreus Sladen, 1919:28
O NOMENCLATURE. B. strenuus Cresson (1863) is a junior secondary homonym in Bombus of Apis strenuus Harris (1776), and therefore the name B. strenuus Cresson is invalid (ICZN, 1985: Article 57c). For this species, the oldest available name is B. neoboreus, which becomes the valid name. The only publications using the name B. strenuus Cresson since 1947 of which I am aware are by Hurd (1979), Milliron (1973a) and Poole (1996), so this change of valid name is not a serious disruption of common usage.
P.H. WILLIAMS
Bombus (AL) polaris Curtis Arcticus Kirby in Parry, 1824:ccexvi, examined, not of Quenzel in Acerbi, 1802:253 (= B. hyperboreus Sch6nherr) Polaris Curtis in Ross, 1835:1xiii, examined diabolicus Friese, 1911:571 alpiniformis Richards, 1931a:13
© NOMENCLATURE. Loken (1973) used the name B. arcticus Kirby for this species because she consid- ered B. arcticus (Quenzel) to be a nomen oblitum. However, this is not supported by the present Code for a publication of this date (ICZN, 1985: Article 79c(iii)), although it does allow B. arcticus (Quenzel) to be suppressed by use of the Plenary Power. See the comments on B. hyperboreus.
APPLICATION TO ICZN. It is suggested above that, in the interests of stability ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to suppress B. arcticus (Quenzel), the unused senior synonym (ICZN, 1985: Article 79) of B. hyperboreus. This would free B. arcticus Kirby from junior primary homonymy with B. arcticus (Quenzel) (ICZN, 1985: Article 57b), so that it would become the valid name for this species (see the comments on B. hyperboreus). However, although the name B. arcticus Kirby has been in use for this species (e.g. Loken, 1973; Sakagami, 1976; Svensson & Lundberg, 1977; Reinig, 1981), the more frequently used name has been B. polaris (e.g. Milliron & Oliver, 1966; Milliron, 1973a: K.W. Richards, 1973; Hurd, 1979; Pekkarinen, 1979; Pekkarinen ef al., 1981; Rasmont, 1983; Pekkarinen & Teris, 1993). In the interests of stability, the application to ICZN might be extended to suppress B. arcticus Kirby, in order to conserve the current usage of B. polaris as the valid name.
Bombus (AI.) alpinus (Linnaeus) alpina (Linnaeus, 1758:579 [Apis]) examined
CHECKLIST OF BUMBLE BEES
Subterraneibombus Skorikov, 1938a:145, unjustified emendation
Bombus (St.) melanurus Lepeletier melanurus Lepeletier, 1836:469, examined ?difficillimus Skorikov, 1912:609, examined subdistinctus Richards, 1928a:333, examined
@ TAXONOMIC sTaTUS. Several of these nominal taxa have been treated as separate species and at least B. difficillimus may indeed prove to be a separate species. However, aside from differences in colour pattern, they are closely similar in morphology with a range of variation (Williams, 1991). Until more evi- dence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (St.) fragrans (Pallas) fragrans (Pallas, 1771:474 [Apis]) mongol Skorikoy, 1912:607, examined ?charharensis Yasumatsu, 1940:94 (provisional synonym)
@ TAXONOMIC STATUS. This taxon is interpreted here in the broadest sense, to include a complex of poorly-known taxa (Williams, 1991). More evidence is awaited.
Bombus (St.) amurensis Radoszkowski Amurensis Radoszkowski, 1862:590, examined
@ TAXONOMIC sTaTUS. I have seen no males of this species and its precise relationships remain un- clear.
Bombus (St.) fedtschenkoi Morawitz Fedtschenkoi Morawitz in Fedtschenko, 1875:5
Bombus (St.) personatus Smith personatus Smith, 1879:132, examined Roborowskyi Morawitz, 1886:197, examined
Bombus (St.) subterraneus (Linnaeus) fubterranea (Linnaeus, 1758:579 [Apis]) examined Nemorum (Scopoli, 1763:307 [Apis])
? faltuum (Panzer, 1801(75):21 [Apis])
O NOMENCLATURE. Loken (1984) interpreted B. saltuum as being conspecific with B. barbutellus (see the comments on B. barbutellus). Warncke (1986) interpreted B. saltuum as having been described from a male (presumably because the antennae were de- scribed as rather long) conspecific with B. subterraneus. No type specimen is known. The de-
121
scription of the anterior part of the gaster of B. saltuum as ashen and the middle part as nearly bald is perhaps slightly closer to B. subterraneus, because although both species may have gastral tergum I with pale hair and terga I-III sparsely haired, this hair is much shorter on B. subterraneus and the posterior fringing hairs of the terga often appear greyish-white.
If B. saltuum were not accepted as most likely to be conspecific with B. subterraneus, then further action would be required. If both interpretations were con- sidered to remain supportable, then it might be considered appropriate (ICZN, 1985: Article 75b) to designate a specimen of B. subterraneus as neotype of Apis saltuum in order to conserve the current usage of B. barbutellus. Alternatively, if B. saltuum were considered more likely to be conspecific with B. barbutellus, then B. saltuum would become the oldest available name for that species, even though the name has not been used in the last 50 years. In the interests of stability (ICZN, 1985: Article 23b), an application could then be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79).
COMMENT. This species has been introduced into New Zealand (e.g. Gurr, 1957; Macfarlane & Gurr, 1995).
Bombus (St.) distinguendus Morawitz nemorum (Fabricius, 1775:382 [Apis]) not of Scopoli, 1763:307 (= B. subterraneus (Linnaeus)), not of Fabricius, 1775:380 (?= B. bohemicus Seidl) elegans Seidl, 1837:67 distinguendus Morawitz, 1869:32
O NOMENCLATURE. The nameB. elegans has been applied to several taxa by different authors. Tkalcit (1969:901-903) reasoned that Seidl had originally described B. elegans from an individual of the species that has more recently been known by the name B. distinguendus, although Seidl’s original type is lost. According toTkalcti, a specimen of B. mesomelas may then have been substituted as the type, but now this cannot be found either. Any remaining confusion could be resolved by the designation of an appropriate neotype. See the comments on B. mesomelas.
€ APPLICATIONTOICZN. AlthoughB. elegans may be the oldest available name for the present interpreta- tion of this species, the name B. distinguendus has been in common use for the species since 1947 (e.g. Tkalcti, 1969, 1974a; Loken, 1973; Alford, 1975; Delmas, 1976; Sakagami, 1976; Pekkarinen, 1979; Reinig, 1981; Pekkarinen er al/., 1981; Rasmont, 1983; Pekkarinen & Teréas, 1993; Rasmont ef al., 1995). I know of no publications using the name B. elegans for this taxon (only for B. mesomelas Gerstaecker as a misidentification) since 1947. It is suggested that, in
122
the interests of stability (ICZN, 1985: Article 23b), and to prevent confusion with B. mesomelas, an appli- cation be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79) (see the comments on B. muscorum).
Bombus (St.) appositus Cresson appositus Cresson, 1878:183
Bombus (St.) borealis Kirby borealis Kirby, 1837:272
Subgenus ALPIGENOBOMBUS Skorikov
Alpigenobombus Skorikov, 1914a:128, type-species Alpigenobombus pulcherrimus Skorikov (= Bombus kashmirensis Friese) by subsequent designation of Williams, 1991:65
Bombus (Mastrucatobombus) Kriger, 1917:66, type-spe- cies Bombus mastrucatus Gerstaecker (= Bombus wurflenii Radoszkowski) by monotypy
Bombus (Alpigenobombus) Frison, 1927:64
[Nobilibombus Skorikov, 1933a:62, published without fixa- tion of type-species]
[Bombus (Nobilibombus) Bischoff, 1936:12, type-species Nobilibombus morawitziides Skorikov (= Bombus nobilis Friese) by monotypy, published as a junior synonym]
Alpigenibombus Skorikov, 1938b:1, unjustified emenda- tion
[Pyrobombus (Nobilibombus) Milliron, 1961:54, type-spe- cies Bombus nobilis Friese (cited as Bombus nobilis Skorikov) by original designation, published as a junior synonym]
Bombus (Nobilibombus) Richards, 1968:222, type-species Bombus nobilis Friese by original designation (see Williams, 1991)
[Alpegenobombus Wang, 1979:188, incorrect subsequent spelling]
@ TAXONOMIC sTATUS. Richards (1968) treated Alpigenobombus and Nobilibombus as separate subgenera. Following Bischoff (1936), I have treated them as a single subgenus Alpigenobombus (Williams, 1991), for which the evidence for monophyly is strong (Williams, 1995).
P.H. WILLIAMS
Bombus (Ag.) kashmirensis Friese kashmirensis Friese, 1909[September, Tkalcii, 1974b]:673 examined stramineus Friese, 1909[September, Tkalct, 1974b]:673 tetrachromus Cockerell, 1909[November, Tkalci, 1974b]:397, examined pulcherrimus (Skorikov, 1914a:128 [Alpigenobombus])
@ TAXONOMIC sTaTUS. Several of these nominal taxa have been treated as separate species. However, aside from differences in colour pattern, they are closely similar in morphology with a range of varia- tion (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
O NOMENCLATURE. Tkalcti (1974b) first regarded B. kashmirensis and B. stramineus as conspecific and, following the Principle of First Reviser (ICZN, 1985: Article 24), chose B. kashmirensis as the name for the species.
Bombus (Ag.) wurflenii Radoszkowski Wurflenii Radoszkowski, 1859:482, examined [Wurfleini Radoszkowski, 1877b:191, incorrect subsequent spelling] mastrucatus Gerstaecker, 1869:326, examined alpigenus Morawitz, 1874:132
Bombus (Ag.) nobilis Friese ?validus Friese, 1905:510, examined (provisional syno- nym) nobilis Friese, 1905:513 ?sikkimi Friese, 1918:82, examined (provisional syno-
nym)
[morawitziides Skorikov, 1922a:159, published without description]
[moravitziides Skorikov, 1931:203, published without de- scription]
?morawitziides (Skorikov, 1933a:62 [Nobilibombus]) ex- amined (provisional synonym)
Dxizangensis Wang, 1979:188, examined (provisional syno- nym)
chayaensis Wang, 1979:189, examined, new synonym
@ TAXONOMIC STATUS. Several of these nominal taxa have been treated as separate species.
Friese (1905) described B. validus initially (p. 510) as having a quadrate malar area and untoothed mandi- bles, but went on (p. 517) to place it within the mastrucatus (= B. wurflenit) group, which he charac- terised as having a short malar area and toothed mandibles. Tkalcii (1987) designated as lectotype of B. validus a female with a quadrate malar area and multi-toothed mandibles. He also synonymised B. morawitziides with B. validus.
The type specimens of B. nobilis have also been in some doubt (Richards, 1968). In the same publication
CHECKLIST OF BUMBLE BEES
as the description of B. validus, Friese (1905) de- scribed the female of B. nobilis as having a‘quadratisch’ malar area and 4—5 teeth on the mandible (even though he placed it [p. 519] ina group withB. lapidarius).The original description lists several females (particularly from Sichuan), but the only putative type female that I have been able to examine (although it carries no Friese ‘type’ label) is in the Berlin museum collection and is a specimen of B. friseanus labelled ‘Kashgar’ (this locality is outside the known distribution range of either B. nobilis or B. friseanus). The specimen does not match the original description of the mandibles of B. nobilis and so cannot be considered a valid syntype. Nonetheless, the identity of B. nobilis is clear from the original description, so the designation of a neotype is not justified (ICZN, 1985: Article 75b).
B. chayaensis appears to me to be very closely similar to the yellow banded B. nobilis (in the strict sense) and I am unaware of any reason to treat them as separate species.
B. nobilis is interpreted here in the broadest sense, to include a complex of morphologically closely simi- lar taxa (Williams, 1991). At least some of the taxa included may prove to be separate species from B. nobilis. The most obvious variation is in the colour of the pale thoracic bands, which may be yellow (B. nobilis), yellow-white (B. sikkimi), gey-white (B. morawitziides), or almost completely replaced by black (B. validus). However, aside from these differences in colour pattern, they are similar in morphology with a range of variation. Until more evidence to the contrary is available from critical studies of patterns of varia- tion, I shall treat them as parts of a single variable species.
O NOMENCLATURE. Following the Principle of First Reviser (ICZN, 1985: Article 24), and as the first author to regard these taxa as conspecific, I select the name B. nobilis as the valid name in preference to B. validus from the available names for this species from Friese (1905).
Bombus (Ag.) genalis Friese
genalis Friese, 1918:84, examined
@ TAXONOMIC sTaTUS. Ihave seen no males of this species and its precise relationships remain unclear.
Bombus (Ag.) grahami (Frison) grahami (Frison, 1933:334 [Bremus])
Bombus (Ag.) breviceps Smith nasutus Smith, 1852a:44, examined breviceps Smith, 1852a:44, examined dentatus Handlirsch, 1888:227 simulus Gribodo, 1892:114, examined
123
orichalceus Friese, 1916:107
rufocognitus Cockerell, 1922:4, examined
pretiosus Bischoff, 1936:11, examined, not of Friese, 1911:571 (= B. polaris Curtis)
2angustus Chiu, 1948:59 (provisional synonym)
bischoffiellus (Tkalct, 1977:224 [Alpigenobombus]) re- placement name for pretiosus Bischoff, 1936:11
@ TAXONOMIC sTATUS. Several of these nominal taxa have been treated as separate species. At least B. dentatus [Himalaya] and B. angustus [Taiwan] may prove to be separate species (e.g. Tkalct, 1968), 1989). However, aside from differences in colour pattern, they are similar in morphology with a range of varia- tion (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
O NOMENCLATURE. Tkalcii (1968d) first regarded B. nasutus and B. breviceps as likely to be conspecific and, following the Principle of First Reviser (ICZN, 1985: Article 24), chose B. breviceps as the name for the species.
Subgenus PYROBOMBUS Dalla Torre
Bombus (Pyrobombus) Dalla Torre, 1880:40, type-species Apis hypnorum Linnaeus (= Bombus hypnorum (Linnaeus)) by monotypy
Bombus (Poecilobombus) Dalla Torre, 1882:23, type-spe- cies Bombus sitkensis Nylander by subsequent designation of Sandhouse, 1943:589
[Bombus (Pyrrhobombus) Dalla Torre, 1882:28, incorrect subsequent spelling]
Bombus (Pyrrhobombus) Dalla Torre, 1896:503, unjusti- fied emendation
Bombus (Pratobombus) Vogt, 1911:49, type-species Apis pratorum Linnaeus (= Bombus pratorum (Linnaeus)) by subsequent designation of Frison, 1927:67
[Bombus (Pratibombus) Ball, 1914:78, incorrect subse- quent spelling]
Bombus (Anodontobombus ) Kriiger, 1917:61,65 (proposed as a section name but stated by Milliron, 1961:53, to be synonymous with his concept of the subgenus Pyrobombus Dalla Torre), type-species Apis hypnorum Linnaeus (= Bombus hypnorum (Linnaeus)) by subse- quent designation of Williams, 1991:69
124
Bombus (Uncobombus) Vogt in Kriiger, 1917:65 (pro- posed as a group name but stated by Milliron, 1961:53, to correspond to his concept of Pyrobombus Dalla Torre), type-speciesApis hypnorum Linnaeus (=Bombus hypnorum (Linnaeus)) by subsequent designation of Williams, 1991:69
Bombus (Lapponicobombus) Quilis-Pérez, 1927:19, type- speciesApis lapponica Fabricius (=Bombus lapponicus (Fabricius)) by subsequent designation of Milliron, 1961:58
[Bombus (Hypnorubombus) Quilis-Pérez, 1927:19, incor- rect original spelling]
[Bombus (Laponicobombus) Quilis-Pérez, 1927:63, incor- rect original spelling]
Bombus (Hypnorobombus) Quilis-Pérez, 1927:97, type- species Apis hypnorum Linnaeus (= Bombus hypnorum (Linnaeus)) by monotypy
Pratibombus Skorikoy, 1938b:1, unjustified emendation
COMMENT. This is the largest subgenus of Bombus. The highest richness of Pyrobombus species occurs in the New World (there are no species known from south of Panama), although the earliest-diverging species within the subgenus appear to be in the Old World (Williams, 1991). This is the opposite pattern to that shown by the next-largest subgenus (of social para- sites), Psithyrus (see the comments on the subgenus Psithyrus).
Bombus (Pr.) abnormis (Tkalci) abnormis (Tkalcii, 1968a:33 [Pyrobombus]) examined
Bombus (Pr.) hypnorum (Linnaeus)
Hypnorum (Linnaeus, 1758:579 [Apis]) examined
leucopygus Wlliger, 1806:172
calidus Erichson in Middendorff, 1851:65
fletcheri Richards, 1934:90, examined
insularis Sakagami & Ishikawa, 1969:180, not of Smith, 1861:155 (= B. insularis (Smith))
Koropokkrus Sakagami & Ishikawa, 1972:610, replace- ment name for insularis Sakagami & Ishikawa, 1969:180
@ TAXONOMIC sTaTuS. B. hypnorum is a broadly distributed species with a fairly easily recognised brown-black-white colour pattern (e.g. Reinig, 1939; Williams, 1991). Recently, Starr (1992) has described what appears to be a divergent, brownish-black or- ange-tailed colour form from a disjunct peripheral population on the island of Taiwan.
Bombus (Pr.) perplexus Cresson perplexus Cresson, 1863:91
Bombus (Pr.) haematurus Kriechbaumer haematurus Kriechbaumer, 1870:157
Bombus (Pr.) subtypicus (Skorikov) leucopygus Morawitz in Fedtschenko, 1875:3, not of Illiger,
P.H. WILLIAMS
1806:172 (= B. hypnorum (Linnaeus))
[leucopygos (Skorikov, 1914b:294 [Pratobombus]) incor- rect subsequent spelling]
subtypicus (Skorikoy, 1914b:294 [Pratobombus]) exam- ined
leucurus Bischoff & Hedicke, 1931:391, replacement name for leucopygus Morawitz in Fedtschenko, 1875:3
kohistanensis (Tkalcti, 1989:49 [Pyrobombus]) examined
Bombus (Pr.) mirus (Tkalcii) mirus (Tkalcu, 1968a:37 [Pyrobombus]) examined ?tibetanus Friese, 1913:86, examined, not of Morawitz, 1886:202 (= B. tibetanus (Morawitz))
Bombus (Pr.) lemniscatus Skorikov lemniscatus Skorikov, 1912:607, examined flavopilosus Friese, 1918:84, examined peralpinus Richards, 1930:646, examined
Bombus (Pr.) lepidus Skorikoy lepidus Skorikoy, 1912:606, examined genitalis Friese, 1913:85, examined tetrachromus Friese, 1918:85, examined, not of Cockerell, 1909:397 (= B. kashmirensis Friese) ?yuennanicola Bischoff, 1936:7, examined
Bombus (Pr.) infirmus (Tkalcit) leucurus Bischoff, 1936:8, examined, not of Bischoff & Hedicke, 1931:391 (= B. subtypicus (Skorikov)) infirmus (Tkalct, 1968a:24 [Pyrobombus]) replacement name for /eucurus Bischoff, 1936:8
Bombus (Pr.) parthenius Richards parthenius Richards, 1934[14April, Williams & Cameron, 1993]:89, examined ?sonani (Frison, 1934[30 April, Williams & Cameron, 1993]:175 [Bremus]) examined ?infrequens (Tkalctt, 1989:56 [Pyrobombus]) examined (provisional synonym)
@ TAXONOMIC STATUS. Several of these nominal taxa have been treated as separate species. At least B. sonani [Taiwan] and B. infrequens [northern Burma to southern China] may prove to be separate species. However, aside from differences in colour pattern, they are closely similar in morphology (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (Pr.) luteipes Richards luteipes Richards, 1934:89, examined ?avanus (Skorikoy, 1938b:2 [Pratibombus]) (provisional synonym) signifer (Tkalctt, 1989:52 [Pyrobombus]), examined
@ TAXONOMIC STATUS. The identity of B. avanus
CHECKLIST OF BUMBLE BEES
is in doubt because the type cannot be found. The description of the colour pattern resembles B. parthenius and B. luteipes, and the description of the longer than usual squama (= gonostylus) and the half- crooked apex of the sagitta (= penis valve) appear to be closely similar to B. luteipes, so these taxa are very likely to be conspecific. More evidence is awaited.
Bombus (Pr.) flavescens Smith flavescens Smith, 1852a:45, examined mearnsi Ashmead, 1905:959 baguionensis Cockerell, 1920:631, new synonym tahanensis Pendlebury, 1923:65, examined ?rufoflavus Pendlebury, 1923:66, examined (provisional synonym)
@ TAXONOMIC sTATUS. Several of these nominal taxa have been treated as separate species. B. rufoflavus [Peninsular Malaysia] and B. baguionensis [Philip- pines] are particularly distinct in colour pattern. They may prove to be separate species, but from the material available from a few sites, they appear to me to be closely similarin morphology to B. flavescens (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (Pr.) rotundiceps Friese rotundiceps Friese, 1916:108, examined montivolanoides Sakagami & Yoshikawa, 1961:431 Shillongensis (Tkalct, 1974b:334 [Pyrobombus]) exam- ined
Bombus (Pr.) beaticola (Tkalcii) beaticola (Tkalcti, 1968a:28 [Pyrobombus]) examined
Bombus (Pr.) picipes Richards flavus Friese, 1905:517, examined, not of Pérez, 1884:265 (= B. campestris (Panzer)) picipes Richards, 1934:90, examined klapperichi Pittioni, 1949:266, examined ?nikiforuki Tkalct, 1961b:354 (provisional synonym)
@ TAXONOMIC STATUS. B. nikiforuki was described from a single worker from Qinghai. From the descrip- tion, it appears to be closely similar to B. picipes and is likely to be conspecific.
O NOMENCLATURE. With Psithyrus regarded as being a subgenus of the genus Bombus (Williams, 1991, 1995), B. pratorum subsp. flavus Friese (1905) becomes a junior secondary homonym in Bombus of Psithyrus campestris var. flavus Pérez (1884) (deemed to be subspecific, see ICZN, 1985: Article 45g(ii)), and therefore the name B. flavus Friese is invalid (ICZN, 1985: Article 57c). For this species, the oldest available name of which I am aware is B. parthenius
125
var. picipes Richards, 1934 (deemed to be subspecific, see ICZN, 1985: Article 45g(11)), which becomes the valid name, B. picipes. The only publications using the name B. flavus Friese since 1947 of which I am aware are by Sakagami (1972), Ito (1993) and Yao & Luo (1997), so this change of valid name is not a serious disruption of common usage.
Bombus (Pr.) ardens Smith ardens Smith, 1879:133, examined andreae Friese, 1910:405, examined
Bombus (Pr.) modestus Eversmann modestus Eversmann, 1852:134 Baikalensis Radoszkowski, 1877b:203 nymphae Skorikoy, 1910b:409 eversmanni Skorikoy, 1910c:581, not infrasubspecific af- ter Skorikov, 1922a:149
Bombus (Pr.) cingulatus Wahlberg cingulatus Wahlberg, 1854:208
COMMENT. The distribution of B. cingulatus accord- ing to Reinig (1939) is shown in Fig. 10.
?Bombus (Pr.) oceanicus Friese oceanicus Friese, 1909:675, examined oceanicus Friese & Wagner, 1910:52, redescribed
@ TAXONOMICSTATUS. B. oceanicus is known only from the Kurile Islands. A particularly close relation- ship with the otherwise broadly distributed B. cingulatus (absent from the Kuriles, but present in Kamchatka, Reinig, 1939; Ito & Sakagami, 1980; Fig. 10) has been suggested by Ito & Sakagami (1980) and it is possible that they are conspecific. More evidence is awaited.
Bombus (Pr.) brodmannicus Vogt Brodmannicus Vogt, 1909:49, examined
Bombus (Pr.) pratorum (Linnaeus) pratorum (Linnaeus, 1761:424 [Apis]) examined
COMMENT. This species was deliberately introduced into Sydney, Australia, although it is not known to have persisted (Oliff, 1895). Until the twentieth cen- tury, B. pratorum was not known from Ireland, where it is now well established (see references in Alford, 1975, 1980) (see comments on B. monticola).
Bombus (Pr.) jonellus (Kirby) Jonella (Kirby, 1802:338 [Apis]) examined alboanalis Franklin, 1913:385
126
@ TAXONOMIC status. B. alboanalis has been regarded both as a separate species (Franklin, 1913; Frison, 1927) and as conspecific with either B. frigidus (Burks, 1951; Hurd, 1979; Poole, 1996) or B. jonellus (Williams, 1991 [as B. jonellus from western Canada]; Scholl et al., 1995).
Recently, Scholl et al. (1995) concluded from stud- ies of enzyme mobility morphs that whereas B. alboanalis and B. frigidus have separate gene pools, in contrast, B. alboanalis and B. jonellus show a low level of genetic differentiation. They also noted the lack of colour gradation between sympatric B. alboanalis and B. frigidus.
From the limited amount of material I have exam- ined, I believe that B. alboanalis and B. jonellus are morphologically closely similar. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (Pr.) pyrenaeus Pérez pyrenaeus Pérez, [1880, see Baker, 1996d:300]:127, not of Lepeletier, 1832:375 (= B. rupestris (Fabricius)) tenuifasciatus Vogt, 1909:49 [pyreneus Pagliano, 1995:23, incorrect subsequent spell- ing]
O NOMENCLATURE. With Psithyrus regarded as being a subgenus of the genus Bombus (Williams, 1991, 1995), B. pyrenaeus Pérez (1880) becomes a junior secondary homonym in Bombus of Psithyrus pyrenaeus Lepeletier (1832), and therefore the name B. pyrenaeus Pérez is invalid (ICZN, 1985: Article 57c). The next available name, tenuifasciatus, was used by Vogt (1909) for individuals with particular colour patterns from both B. pyrenaeus Pérez and B. sichelii. The choice of which of these two homonyms should have precedence depends on the Principle of the First Reviser (ICZN, 1985: Article 24). As far as I have been able to discover, Tkalciti (1973:266) is the first author to have recognised this problem. He recog- nised precedence for B. pyrenaeus ssp. tenuifasciatus Vogt. Consequently, the oldest available name for this species, and therefore the valid name, is B. tenuifasciatus.
& APPLICATION TO ICZN. Although B. tenuifasciatus is the oldest available name for this species, the name B. pyrenaeus has been in common use for the species since 1947 (e.g. Krusemen, 1958; Tkalct, 1969, 1973, 1975; Reinig, 1972, 1981; Delmas, 1976; Rasmont, 1983; Ornosa, 1986; Williams, 1991; Rasmont ef al., 1995). It is suggested that, in the interests of stability, an application be made to ICZN to use its Plenary Power to suppress the senior homo- nym (ICZN, 1985: Article 79) (see the comments on B. muscorum). However, the consequence of this action
P.H. WILLIAMS
would be that pyrenaeus (Lepeletier) would no longer be available for a subspecies of B. rupestris.
Bombus (Pr.) biroi Vogt biroi Vogt, 1911:51, examined nursei Friese, 1918:84, examined Yagnatus Skorikov, 1933b:248, examined, not of Skorikov, 1912:97 (= B. monticola Smith) ?kotzschi Reinig, 1940:227, examined
@ TAXONOMIC sTATUS. Several of these nominal taxa have been treated as separate species. However, aside from differences in colour pattern, they are closely similar in morphology with a range of varia- tion (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (Pr.) frigidus Smith frigidus Smith, 1854:399, examined
Bombus (Pr.) sandersoni Franklin sandersoni Franklin, 1913:353
Bombus (Pr.) flavifrons Cresson pleuralis Nylander, 1848:231, examined flavifrons Cresson, 1863:105, new synonym
@ TAXONOMIC STATUS. B. pleuralis and B. flavifrons were regarded as separate species by Franklin (1913), and Poole (1996) also lists them as separate species, without explanation. In my opinion, the lectotype of B. pleuralis designated by Milliron (1960:95) is an individual of the dark form of B. flavifrons (see descriptions of variation by e.g. Stephen, 1957; Thorp et al., 1983). See also the comments on B. mixtus.
O NOMENCLATURE. B. pleuralisis the oldest avail- able name for this species.
€ APPLICATION TOICZN. Although B. pleuralis is the oldest available name for the present interpretation of this species, the name B. flavifrons has been in common use for the species since 1947 (e.g. Stephen, 1957; Thorp, 1969, 1970; Plowright & Stephen, 1973; Macior, 1975; Sakagami, 1976; Hurd, 1979; Plowright & Owen, 1980; Thorp et al., 1983; Laverty & Harder, 1988). I know of no publications using the name B. pleuralis since 1947, apart from the list by Poole (1996). It is suggested that, in the interests of stability (ICZN, 1985: Article 23b), an application be made to ICZN to use its Plenary Power to suppress the unused senior synonym (ICZN, 1985: Article 79) (see the
rn
CHECKLIST OF BUMBLE BEES
comments on B. muscorum). However, the conse- quence of this action would be that pleuralis would no longer be available for a species or for a subspecies of B. flavifrons.
Bombus (Pr.) centralis Cresson centralis Cresson, 1864:41
Bombus (Pr.) vandykei (Frison) vandykei (Frison, 1927:375 [Bremus]) cascadensis (Milliron, 1970a:382 [Pyrobombus])
Bombus (Pr.) caliginosus (Frison) caliginosus (Frison, 1927:376 [Bremus])
Bombus (Pr.) vagans Smith vagans Smith, 1854:399, examined
Bombus (Pr.) mixtus Cresson® Praticola Kirby, 1837:274 mixtus Cresson, 1878:186, not of Kriechbaumer, 1870:160 (= B. maxillosus Klug), new synonym
@ TAXONOMICSTATUS. The identity of B. praticola has remained uncertain (e.g. Cresson, 1863; Franklin, 1913). Recently, Poole (1996) has listed B. praticola, B. mixtus and B. flavifrons as separate species without explanation.
Although I know of no type material, Kirby pro- vided a description of B. praticola from northern Canada (latitude 65° North) with a colour pattern (including anterior half of abdomen yellow, posterior ferruginous) that for individuals from this area is most likely to be conspecific either with B. mixtus (some individuals have few black hairs on gastral terga II- III), or with B. flavifrons (which has terga V—VI black, although this is not always apparent from the dorsal view). In his original description of B. flavifrons, Cresson (1863) conceded that this might be the same species as Kirby’s B. praticola, and he went on to write (p. 106) that he had not yet identified B. praticola. Franklin (1913:371) wrote that he had ‘been unable to decide whether the original description of B. praticolus [sic] referred to this species [B. flavifrons] or to the colour variant of pleuralis [intermediate colour pat- terns between B. flavifrons and B. pleuralis].’ Milliron (1971:42) subsequently listed Pyrobombus praticola flavifrons (Cr.) as a member of his ‘Praticola Group’.
However, here I follow R. Miller (in Jitt.), who believes that the original material was more likely to have been of the species that has come to be known as B. mixtus. See the comments on B. flavifrons.
O NOMENCLATURE. B. praticola is probably the oldest available name for this species. Any remaining
127
confusion could be resolved by the designation of an appropriate neotype (e.g. see the comments on B. subterraneus).
© APPLICATION TOICZN. Although B. praticola is probably the oldest available name for this species, the name B. mixtus has been in common use for the species since 1947 (e.g. Stephen, 1957; Thorp, 1970; Plowright & Stephen, 1973; K. W. Richards, 1973; Macior, 1975; Sakagami, 1976; Hurd, 1979; Plowright & Owen, 1980; Thorp et al., 1983; Laverty & Harder, 1988; Macfarlane et al., 1994). It is suggested that, in the interests of stability, an application be made to ICZN to use its Plenary Power to suppress the senior synonym and homonym (ICZN, 1985: Article 79) (see the comments on B. muscorum). However, the conse- quence of this action would be that mixtus (Kriech- baumer) would no longer be available for a subspecies of B. maxillosus.
Bombus (Pr.) sitkensis Nylander Sitkensis Nylander, 1848:235
Bombus (Pr.) melanopygus Nylander melanopyge Nylander, 1848:236 Edwardsii Cresson, 1878:184 melampygus Handlirsch, 1888:231, unjustified emenda- tion [melanopygus Viereck, 1904:99, incorrect subsequent spell- ing] melanopygus Franklin, 1913:334, justified emendation @ TAXONOMIC STATUS. B. melanopygus and B. edwardsii were shown by Owen & Plowright (1980) to differ principally by a single allele controlling the colour of the pubescence on gastral terga II-III. There can be little doubt that they are conspecific.
Bombus (Pr.) lapponicus (Fabricius) lapponica (Fabricius, 1793:318 [Apis]) ?sylvicola Kirby, 1837:272 zhaosu Wang, 1985:162, examined, new synonym
@ TAXONOMIC sTATUS. B. sylvicola is morpho- logically closely similar to B. lapponicus, and it has been suggested repeatedly that they may be conspecific (e.g. Sladen, 1919; Skorikoy, 1922a, 1937; Pittioni, 1942, 1943; Thorp, 1962; Thorp et al., 1983).
B. zhaosu was described from material from Xinjiang, China, and is closely similar toB. lapponicus.
These three nominal taxa have been treated as sepa- rate species. However, aside from differences in colour pattern, they are closely similar in morphology. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
128
?Bombus (Pr.) monticola Smith
montanus Smith, 1844:549, not of Lepeletier, 1836:463 (= B. ruderarius (Miiller))
monticola Smith, 1849:lx, replacement name formontanus Smith, 1844:549
lugubris Sparre-Schneider, 1909:155, not of Kriechbaumer, 1870:159 (= B. maxillosus (Klug))
scandinavicus Friese, 1912:684, replacement name for lugubris Sparre-Schneider, 1909:255
@ TAXONOMIC STATUS. B. scandinavicus (= B. monticola) and B. lapponicus are names that were applied initially to two colour forms in Scandinavia.
Lgken (1973) reported that these two taxa overlap narrowly in distribution and intergrade. However, they have been found to differ consistently (for samples analysed) in the composition of cephalic secretions (Bergstr6m & Svensson, 1973; Svensson & Bergstré6m, 1977). Svensson (1973, 1979) also described subtle differences in morphological characters, although other morphological studies by Lgken (1973) and Pekkarinen (1979) found no distinct differences. Pekkarinen (1982, in litt.) now believes that they are separate species.
It remains possible that there is a hybrid zone where the colour forms intergrade, with some gene flow. In this case, depending on the species concept embraced, these taxa might be considered conspecific (see the com- ments on B. ruderatus). Until further evidence is avail- able, I shall continue to treat them as separate species.
COMMENT. Until the twentieth century B. monticola was not known from Ireland, where it is now estab- lished (see references in Alford, 1975, 1980) (see comments on B. pratorum).
Bombus (Pr.) bimaculatus Cresson bimaculatus Cresson, 1863:92
Bombus (Pr.) impatiens Cresson impatiens Cresson, 1863:90
Bombus (Pr.) vosnesenskii Radoszkowski Vosnesenskii Radoszkowski, 1862:589
Bombus (Pr.) bifarius Cresson bifarius Cresson, 1878:185 andamanus Gribodo, 1882:268, examined fernaldi Franklin, 1911:157, not a replacement name
@ TAXONOMIC sTaTUS. B. andamanus was de- scribed as originating from ‘Andaman’ (= Andaman Islands, Indian Ocean), but appears to be a mislabelled queen of B. bifarius from western North America (Tkalcti, 1966). I have examined this specimen and agree with this identification (i.e. contrary to Richards, 1929, itis nota species of the subgenus Bombus s. str.).
P.H. WILLIAMS
Bombus (Pr.) huntii Greene Huntii Greene, 1860:172
Bombus (Pr.) ternarius Say ternarius Say, 1837:414 ornatus Smith, 1854:398, examined
Bombus (Pr.) ephippiatus Say ephippiatus Say, 1837:414 formosus Smith, 1854:403, examined lateralis Smith, 1879:134, examined ?wilmatte Cockerell, 1912:21, examined ?alboniger Franklin, 1915:409, examined folsomi (Frison, 1923:322 [Bremus]) examined
@ TAXONOMIC STATUS. B. wilmattae, B. alboniger and B. ephippiatus have been regarded both as conspecific and as separate species. Recently, B. wilmattae and B. ephippiatus were regarded as separate species by Labougle ef al. (1985) and Labougle (1990), who described diagnostic charac- ters of colour pattern and morphology. However, D. Yanega (in litt.) and G. Chavarria (pers. com.) believe that all of these nominal taxa are part of the wide- spread and variable B. ephippiatus. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
B. folsomi was described as originating from “Kina Bala / N. Borneo’ (= Gunung Kinabalu, Sabah), but appears to be a mislabelled queen of B. ephippiatus, probably from Costa Rica or Panama (Starr, 1989). I have examined this specimen and agree with this identification.
Subgenus FESTIVOBOMBUS Tkalcii [Atrocinctob.{ombus | Skorikov, 1933b:244, published with- out description] Pyrobombus (Festivobombus) Tkalcu, 1972:26, type-spe- cies Bombus festivus Smith by original designation Bombus (Festivobombus) Williams, 1985b:240
@ TAXONOMIC STATUS. Richards (1968) treated B. atrocinctus (= B. festivus) as a species of the
CHECKLIST OF BUMBLE BEES
subgenus Pyrobombus, even though this required nu- merous exceptions in the diagnosis. I have followed Tkalcii(1972, 1974) in treating Festivobombus and Pyrobombus as separate subgenera (Williams, 1991), because together they do not form a monophyletic group (Williams, 1995).
Subgenus RUFIPEDIBOMBUS Skorikov Rufipedibombus Skorikov, 1922a:156, type-species Bombus rufipes Lepeletier by monotypy Bombus (Rufipedibombus) Richards, 1930:638 Bombus (Rufipedobombus) Kruseman, 1952:102, unjusti- fied emendation
Bombus (Rf) rufipes Lepeletier rufipes Lepeletier, 1836:473 richardsi (Frison, 1930:6 |Bremus])
Bombus (Rf.) eximius Smith eximius Smith, 1852b:47, examined latissimus Friese, 1910:405
129
Bremus pressus Frison (= Bombus pressus (Frison)) by original designation
Bombus (Pressobombus) Kruseman, 1952:102, unjusti- fied emendation
Bombus (Pressibombus) Richards, 1968:217
Bombus (Pe.) pressus (Frison) pressus (Frison, 1935:342 [Bremus])
Subgenus BOMBUS in the strict sense
Bombus (Leucobombus) Dalla Torre, 1880:40, type-spe- cies Apis terrestris Linnaeus (= Bombus terrestris (Linnaeus)) by subsequent designation of Sandhouse, 1943:564
Bombus (Terrestribombus) Vogt, 1911:55, type-species Apis terrestris Linnaeus (= Bombus terrestris (Linnaeus)) by subsequent designation of Frison, 1927:67
Bombus (Bo.) sporadicus Nylander sporadicus Nylander, 1848:233 malaisei Bischoff, 1930:4
Bombus (Bo.) tunicatus Smith tunicatus Smith, 1852a:43, examined vallestris Smith, 1878:8 gilgitensis Cockerell, 1905:223, examined
Bombus (Bo.) franklini (Frison) franklini (Frison, 1921:147 [Bremus])
@ TAXONOMIC STATUS. This species has been treated as conspecific with B. occidentalis (= B. terricola) by Milliron (1971), but has since been shown to be very distinct in morphology by Plowright & Stephen (1980) and Williams (1991), and in enzyme mobilities by Scholl, Thorp & Obrecht (1992).
COMMENT. 8B. franklini has one of the narrowest distributions of any bumble bee species world-wide. All recent specimens have been collected within a 60 mile (38 km) radius of Grants Pass, Oregon (Thorp, 1970; Thorp et al., 1983).
130
Bombus (Bo.) affinis Cresson affinis Cresson, 1863:103
Bombus (Bo.) ignitus Smith ignitus Smith, 1869:207, examined terminalis Smith, 1873:206, examined, not of Smith in Horne, 1870:193 (= B. festivus Smith) japonicus Dalla Torre, 1890:139, replacement name for terminalis Smith, 1873:206
Bombus (Bo.) terrestris (Linnaeus) terreftris (Linnaeus, 1758:578 [Apis]) Audax (Harris, 1776:130 [Apis]) not of Harris, 1776:137 (= Anthophora sp.) canariensis Pérez, 1895:191 maderensis Erlandsson, 1979:191
@ TAXONOMICSTATUS. B. terrestris, B. maderensis and B. canariensis have been regarded both as conspecific and as separate species.
Erlandsson (1979) argued that the dark individuals from the Canary Islands, previously placed within B. terrestris by for example Kriiger (1954, 1956), are a separate species, B. canariensis. Erlandsson also ar- gued that individuals from the island of Madeira, previously placed within B. terrestris by Bischoff (1937), are a separate species, B. maderensis. In both cases the morphological characters used to support these distinctions are not strongly divergent from the broad variation within B. terrestris in the broad sense. Rasmont (1984) regarded these three taxa as separate species, but Pekkarinen & Kaarnama (1994) treated them as conspecific.
Recent work by Estoup et al. (1996) has found that although European mainland populations do not vary significantly among themselves in mitochondrial genes, all island populations studied (from six Medi- terranean islands in addition to B. canariensis) show significant differences from the mainland populations.
Consequently, viewing these three nominal taxa as separate species may be one interpretation, but this appears to depend on adopting a species concept that admits little colour, morphological or genetic varia- tion within a species and regards current geographical isolation as highly indicative. I prefer to regard these taxa as conspecific until further evidence is available.
O NOMENCLATURE. Day (1979) described how none of the admissable syntypes of A. terrestris Linnaeus is in agreement with the current usage of the name.
To reaffirm the traditional usage of this particularly widely used name, a case was made to ICZN by Loken et al. (1994). This sought an Opinion from ICZN (ICZN, 1996) that set aside, by use of its Plenary Power (ICZN, 1985: Articles 78b, 79), the lectotype designation for A. terrestris by Day from application
P.H. WILLIAMS
of the Code (ICZN, 1985), and then designated a neotype (ICZN, 1996: 64) to conserve the traditional usage of the name for even the narrowest concept of the taxon (ICZN, 1985: Article 75).
COMMENT. This species has been introduced into New Zealand (e.g. Gurr, 1957; Macfarlane & Gurr, 1995) (see the comments onB. hortorum, B. ruderatus, and B. subterraneus), Tasmania (Cardale, 1993), and Japan (I. Washitani, in /itt.). It was also apparently introduced into mainland Australia (New South Wales) without persisting (W. Froggatt in Franklin, 1913).
Bombus (Bo.) hypocrita Pérez hypocrita Pérez, 1905:30
Bombus (Bo.) patagiatus Nylander patagiatus Nylander, 1848:234 vasilievi Skorikov, 1913:172
Bombus (Bo.) lucorum (Linnaeus)
lucorum (Linnaeus, 1761:425 [Apis]), examined
2cryptarum (Fabricius, 1775:379 [Apis])
?modestus Cresson, 1863:99, not of Eversmann, 1852:134 (= B. modestus Eversmann)
?moderatus Cresson, 1863:109, replacement name for modestus Cresson, 1863:99
monozonus Friese, 1909:674
?magnus Vogt, 1911:56
2jacobsoni Skorikov, 1912:610, examined
?burjaeticus Kriiger, 1954:277
?florilegus Panfilov, 1956:1334
?reinigi Tkalct, 1974b:322, examined
@ TAXONOMIC sTaTUS. These bees have received particularly close attention by authors describing the minutiae of colour variation, using at least 186 classi- cal names (see the introduction). At least some of these nominal taxa have been regarded as separate species by some authors (e.g. Rasmont, 1983, 1984, 1988; Scholl & Obrecht, 1983; Scholl et al., 1990; Scholl, Thorp & Obrecht, 1992; Rasmont et al., 1995; Amiet, 1996; Ozbek, 1997; Pamilo et al., 1997). In contrast, B. cryptarum and the North American B. moderatus have recently been treated as conspecific with B. lucorum by Poole (1996).
There are conflicting patterns of variation among some characters of these taxa, which are not fully understood (Pekkarinen, 1979; Pamilo ef al., 1984; Amiet, 1996; Pamilo et al., 1997). As far as I can tell from the evidence available at present, separation of the taxa for mapping is still not reliable, at least in some areas of their distribution, and particularly in Central and Eastern Asia, where there appears to be a broad range of variation with some intergradation of character combinations (Williams, 1991). Therefore,
CHECKLIST OF BUMBLE BEES
because complete mapping of separate taxa is not yet possible for me, B. /ucorum is interpreted here in the broadest sense, to include a complex of similar taxa. However, these taxa require more critical work to clarify population patterns of variation and inherit- ance, even in relatively well known areas such as Britain.
O NOMENCLATURE. When Day (1979) came to fix the application of A. terrestris Linnaeus (see the comments on B. terrestris), he had no reason to believe that Linnaeus had not described this taxon from the syntype specimen that was subsequently described as the lectotype (=A. cryptarum Fabricius, see Rasmont, 1988:52, ?= B. lucorum (Linnaeus)). This action brought B. Jucorum (Linnaeus) into subjective junior synonymy with B. terrestris (Linnaeus).
To reaffirm the traditional usage of B. terrestris and B. lucorum, a case was made to ICZN by Loken et al. (1994). This sought an Opinion from ICZN (ICZN, 1996) that set aside, by use of its Plenary Power (ICZN, 1985: Articles 78b, 79), the lectotype designa- tion for A. terrestris by Day from application of the Code (ICZN, 1985), and then designated a neotype (ICZN, 1996: 64) to conserve the traditional usage of B. terrestris and B. lucorum (ICZN, 1985: Article 75).
COMMENT. This species occurs in Iceland, where it has probably been introduced (Prys-Jones et al., 1981) (see the comments on B. hortorum).
Bombus (Bo.) terricola Kirby Terricola Kirby, 1837:273 occidentalis Greene, 1858:12
@ TAXONOMIC sTaTUS. B. terricola and B. occidentalis have been regarded both as conspecific (e.g. Milliron, 1971; Poole, 1996) and as separate species (e.g. Franklin, 1913 [but see p. 239]; Stephen, 1957; Thorp et al., 1983; Scholl et al., 1990).
Many specimens from the north west of North America show a reduction in the extent of the yellow bands on gastral terga I and II, with an expansion of the pale pubescence on tergum IV, and so appear to be intermediate or recombinant individuals. Indeed, Stephen’s (1957:74) figure 4 shows several patterns that could represent a continuum of variation between the two forms. Furthermore, Thorp ef al. (1983: fig. 140a) illustrate individuals of ‘B. occidentalis’ from California that look very similar to eastern B. terricola. In view of this, Stephen’s conclusion that there is no intergradation may result from adopting colour crite- ria (identifying B. terricola in the strict sense either by completely black pubescence of female terga V—VI [p. 15] and male tergum IV [p. 19], or by completely yellow pubescence of tergum II [pp. 19, 71], two character states that do not always occur together, even
131
in the east) that could be considered as essentially arbitrary points on a continuum (see the comments on B. fervidus).
In view of the existence of apparent intermediates between these nominal taxa in at least part of their range, they are treated here as likely to be conspecific. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Subgenus CULLUMANOBOMBUS Vogt
Bombus (Cullumanobombus) Vogt, 1911:57, type-species Apis cullumana Kirby (= Bombus cullumanus (Kirby)) by subsequent designation of Frison, 1927:66
Bremus (Rufocinctobombus) Frison, 1927:78, type-spe- cies Bombus rufocinctus Cresson by monotypy
Cullumanibombus Skorikov, 1938a:145, unjustified emen- dation
Bombus (Cu.) rufocinctus Cresson rufo-cinctus Cresson, 1863:106
Bombus (Cu.) cullumanus (Kirby) Cullumana (Kirby, 1802:359 [Apis]) examined serrisquama Morawitz, 1888:224 Silantjewi Morawitz, 1892:132 apollineus Skorikovy, 19106:412
@ TAXONOMIC sTATUS. Several of these nominal taxa have been treated as separate species. However, aside from differences in colour pattern, they are closely similar in morphology (Panfilov, 1951). Rasmont (1988) has drawn attention to the co-occur- rence of the white-banded B. apollineus with the yellow-banded B. serrisquama in northern Iran, ap- parently without intermediate individuals. But by analogy, it is possible that this colour difference could be the effect of a single allele for pigment (cf. B. melanopygus, see also the comments on B. keriensis). Until more evidence to the contrary is available from critical studies of patterns of varia- tion, I shall treat them as parts of a single variable species.
sy
Bombus (Cu.) unicus Morawitz unicus Morawitz, 1883:235 controversus Skarikov, 1910b: 411
@ TAXONOMIC STATUS. B. unicus is similar to B. cullumanus and could possibly be conspecific. How- ever, the male genitalia appear to be more distinct (Panfilov, 1951) from those of the other taxa tradition- ally considered subspecies of B. cullumanus.
Subgenus OBERTOBOMBUS Reinig Bombus (Obertobombus) Reinig, 1930:107, type-species Bombus oberti Morawitz by monotypy [Obertibombus Skorikoy, 1931:239, incorrect subsequent spelling] Bombus (Obertibombus) Reinig, 1934:167, unjustified emendation
@ TAXONOMIC STATUS. Richards (1968) treated Obertobombus as a synonym of the subgenus Sibiricobombus. I have recognised Obertobombus and Sibiricobombus as separate subgenera, because to- gether they do not form a monophyletic group (Williams, 1995).
Bombus (Ob.) morawitzi Radoszkowski Morawitzi Radoszkowski, 1876:101, examined hydrophthalmus Morawitz, 1883:240, examined
Bombus (Ob.) oberti Morawitz Oberti Morawitz, 1883:238, examined Semenovi Morawitz, 1886:198, examined xionglaris Wang, 1982:432, examined, new synonym duanjiaoris Wang, 1982:444, examined zhadaensis Wang, 1982:444, examined, new synonym
@ TAXONOMIC STATUS. B. xionglaris and B. zha- daensis are closely similar to B. oberti in morphology and in colour pattern. These bees occur at high alti- tudes and are not common in collections (Williams, 1991). However, I know of no reason why these
P.H. WILLIAMS
nominal taxa should not be considered conspecific.
Subgenus MELANOBOMBUS Dalla Torre
Bombus (Melanobombus) Dalla Torre, 1880:40, type-spe- cies Apis lapidaria Linnaeus (= Bombus lapidarius (Linnaeus)) by subsequent designation of Sandhouse, 1943:569
Bombus (Lapidariobombus) Vogt, 1911:58, type-species Apis lapidaria Linnaeus (= Bombus lapidarius (Linnaeus)) by subsequent designation of Sandhouse, 1943:562
Kozlovibombus Skorikov, 1922a:152, type-species Bombus kozlovi Skorikov, 1910b (=Bombus keriensis Morawitz) in the sense of Skorikov, 1922a (based on males = Bombus pyrosoma Morawitz, a misidentification, see Reinig, 1934:169, requiring designation by ICZN), by subsequent fixation of Sandhouse, 1943:561
Bombus (Kozlowibombus) Bischoff, 1936:10, unjustified emendation
Lapidariibombus Skorikov, 1938a:145, unjustified emen- dation
?Bombus (Tanguticobombus) Pittioni, 1939d:201, type- species Bombus tanguticus Morawitz by original designation (provisional synonym)
[Bombus (Lapedariobombus )Esmaili & Rastegar, 1974:52, incorrect subsequent spelling]
Bombus (MI.) tanguticus Morawitz tanguticus Morawitz, 1886:200
@ TAXONOMIC STATUS. Queens of B. tanguticus are morphologicaly very distinctive (discussed in Williams, 1991), so much so that Pittioni (1939d) considered the species warranted a subgenus of its own. The male remains apparently unknown (the spe- cies occurs at high altitudes in Tibet [= Xizang] and is very rare in collections), so that its precise relation- ships are difficult to resolve at present and a separate subgenus seems premature.
Bombus (M1.) simillimus Smith simillimis Smith, 1852b:48, examined [similis Smith, 1854:403, incorrect subsequent spelling] [simillimus Dalla Torre, 1896:548, incorrect subsequent spelling] grossiventris Friese, 1931:303, examined
CHECKLIST OF BUMBLE BEES
oculatus (Frison, 1933:335 [Bremus]) examined tonsus (Skorikov, 1933b:248 [Sibiricobombus]) examined simillimus Williams, 1991:99, justified emendation
Bombus (ML.) richardsiellus (Tkalcii) richardsiellus (Tkalci, 1968a:42 [Pyrobombus]) exam- ined
Bombus (ML.) pyrosoma Morawitz pyrosoma Morawitz, 1890:349, examined pyrrhosoma Dalla Torre, 1896:544, unjustified emenda- tion wutaishanensis (Tkalcit, 1968a:39 [Pyrobombus]) exam- ined
@ TAXONOMICSTATUS. B. pyrosomahas been con- sidered conspecific with B. friseanus (Bischoff, 1936) and has been considered conspecific with B. formosellus, B. friseanus and B. flavothoracicus (= B. miniatus) (Williams, 1991). From a preliminary analy- sis of colour variation, S.-f. Wang and J. Yao report (in litt.) that these taxa appear to remain discrete and are likely to be separate species. More evidence is awaited.
?Bombus (MI.) formosellus (Frison)
formosellus (Frison, 1934:163 [Bremus]) examined
@ TAXONOMIC status. B. formosellus has been considered conspecific with B. pyrosoma, B. friseanus and B. flavothoracicus (= B. miniatus) (Williams, 1991), as a disjunct peripheral population on Taiwan. From a preliminary analysis of colour variation, S.-f. Wang and J. Yao report (in /itt.) that these taxa appear to remain discrete and are likely to be separate species. More evidence is awaited.
?Bombus (ML.) friseanus Skorikov friseanus Skorikov, 1933a:62, examined hénei Bischoff, 1936:10, examined
@ TAXONOMIC STATUS. _ B. friseanus has been con- sidered conspecific with B. pyrosoma (Bischoff, 1936; Tkalcti, 1961b; Sakagami, 1972) and has been consid- ered conspecific with B. pyrosoma, B. formosellus and B. flavothoracicus (= B. miniatus) (Williams, 1991). From a preliminary analysis of colour variation, S.-f. Wang and J. Yao report (in /itt.) that these taxa appear to remain discrete and are likely to be separate species. More evidence is awaited.
?Bombus (ML.) miniatus Bingham flavothoracicus Bingham, 1897:552, examined, not of Hoffer, 1889:49 (= B. campestris (Panzer)) miniatus Bingham, 1897:553, examined eurythorax Wang, 1982:435, examined, new synonym
133 stenothorax Wang, 1982:439, examined, new synonym
@ TAXONOMIC STATUS. B. miniatus has been con- sidered conspecific with B. pyrosoma, B. formosellus and B. friseanus (Williams, 1991).
Evidence of intermediates between B. miniatus and B. friseanus is not strong, but not least because so little material is available from where these taxa occur in close proximity in the eastern Himalaya The few workers and males from this area that I have seen are difficult to assign to either taxon with any confidence, although the queens are closer to the colour pattern of B. miniatus (Williams, 1991). More evidence is awaited.
B. eurythorax and B. stenothorax are closely similar in morphology and colour pattern to B. miniatus. I know of no reason why these nominal taxa should not be considered conspecific.
O NOMENCLATURE. With Psithyrus regarded as being a subgenus of the genus Bombus, B.
flavothoracicus Bingham (1897) becomes a junior
secondary homonym in Bombus of Psithyrus camp- estris var. flavothoracicus Hoffer (1889) (deemed to be subspecific, see ICZN, 1985: Article 45g(ii)), and therefore the name B. flavothoracicus Bingham is invalid (ICZN, 1985: Article 57c).
For this species, the oldest available name is B. miniatus, which becomes the valid name. The only subsequent publications of which I am aware that use the name B. flavothoracicus for this taxon as a species are by Tkalcti (1974b), Wang (1982) and Macior (1990), so this change of valid name is not a serious disruption of common usage.
Bombus (M1.) rufofasciatus Smith
rufo-fasciatus Smith, 1852b:48, examined
Prshewalskyi Morawitz, 1880:342
rufocinctus Morawitz, 1880:343, examined, not of Cresson, 1863:106 (= B. rufocinctus Cresson)
chinensis Dalla Torre, 1890{June 25]:139, replacement name for rufocinctus Morawitz, 1880:343; not of Morawitz, 1890[April 30]:352 (= B. chinensis (Morawitz))
waterstoni Richards, 1934:88, examined
Bombus (MI.) ladakhensis Richards ladakhensis Richards, 1928a:336, examined, not infrasubspecific after Tkalcti, 1974b:335 phariensis Richards, 1930:642, examined, not infrasubspecific after Tkalcti, 1974b:336 variopictus Skorikov, 1933b:248, examined reticulatus Bischoff, 1936:7, examined
Bombus (MI.) semenovianus (Skorikov) semenovianus (Skorikov, 1914a:127 [Lapidariobombus]) examined
134
Bombus (MIL.) incertus Morawitz incertus Morawitz, 1881:229
Bombus (ML.) lapidarius (Linnaeus) Lapidaria (Linnaeus, 1758:579 [Apis]) examined Strenuus (Harris, 1776:131 [Apis]) eriophorus Klug, 1807:265, examined caucasicus Radoszkowski, 1859:482, examined
Bombus (ML.) keriensis Morawitz
keriensis Morawitz, 1886:199, examined
separandus Vogt, 1909:61, examined
kohli Vogt, 1909:61, examined, not of Cockerell, 1906:75 (= B. morio (Swederus))
kozlovi Skorikoy, 1910b:413, replacement name for kohli Vogt, 1909:61
tenellus Friese, 1913:86
[alagesianus (Skorikov, 1922a:152 [Lapidariobombus]) published without description]
alagesianus Reinig, 1930:89
richardsi Reinig, 1935:341, not of Frison, 1930:6 (= B. rufipes Lepeletier)
tibetensis Wang, 1982:439, replacement name forrichardsi Reinig, 1935:341
trilineatus Wang, 1982:441, examined, new synonym
@ TAXONOMIC STATUS. Several of these nominal taxa have been treated as separate species (e.g. Skorikov, 1931), although B. keriensis has also long been considered a broadly-distributed and variable species, including both yellow-banded and white- banded individuals throughout much of its range (Reinig, 1935, 1939; Williams, 1991; Fig. 9).
B. trilineatus is morphologically closely similar to B. keriensis. | know of no reason why these nominal taxa should not be considered conspecific.
Bombus (ML1.) sichelii Radoszkowski
Sichelii Radoszkowski, 1859:481, examined
[Sicheli Radoszkowski, 1877b:213, incorrect subsequent spelling]
tenuifasciatus Vogt, 1909:49, not of Vogt, 1909:49 (= B. pyrenaeus Pérez) after Tkalcii, 1973:266
chinganicus Reinig, 1936:6, not of Reinig, 1936:8 (?= B. bohemicus Seidl)
erzurumensis (Ozbek, 1990:209 [Pyrobombus]) examined, new synonym
@ TAXONOMIC sTaTUS. Until recently, the white- banded form of B. sichelii has been known from west of the Caspian Sea only from the Caucasus (Reinig, 1935). Now that B. erzurumensis (morphologically closely similar to B. sicheli-*
i and with white bands) has been described from Turkey, it could be interpreted as another white-banded, western colour form. By analogy (cf. comments on B. melanopygus), the difference in colour could be the
P.H. WILLIAMS
effect of a single allele for pigment.
Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Subgenus SIBIRICOBOMBUS Vogt
Bombus (Sibiricobombus) Vogt, 1911:60, type-speciesApis sibirica Fabricius (= Bombus sibiricus (Fabricius)) by subsequent designation of Sandhouse, 1943:599
Sibiricibombus Skorikov, 1938a:145, unjustified emenda- tion
[Bombus (Sibericobombus) Kruseman, 1952:101, incor- rect subsequent spelling]
Bombus (Sb.) sibiricus (Fabricius) fibirica (Fabricius, 1781:478 [Apis]) examined flaviventris Friese, 1905:514, examined, new synonym ochrobasis Richards, 1930:655, examined, new synonym
@ TAXONOMIC sTaTuUS. B. sibiricus and B. flavi- ventris have been regarded as separate species. Females of B. flaviventris are morphologically closely similar to those of B. sibiricus, but differ in having the orange pubescence dorsally between the wing bases and on gastral terga IV—VI replaced with black. S.-f. Wang and J. Yao have kindly shown me the male of B. flaviventris, which 1s closely similar in its genitalia to B. sibiricus.
B. ochrobasis appears to differ from B. flaviventris only in the lighter hue of the yellow pubescence of B. ochrobasis.
At present I know of no good biological reason why these three nominal taxa should not be regarded as conspecific. More evidence is awaited.
COMMENT. B. flaviventris has long been placed in the subgenus Subterraneobombus (e.g. Skorikov, 1922a; Richards, 1930, 1968), although the characters of the females (Williams, 1991) and the males (Wang & Yao, unpublished) agree with the species of the subgenus Sibiricobombus.
Bombus (Sb.) obtusus Richards obtusus Richards, 1951:196, examined
CHECKLIST OF BUMBLE BEES
Bombus (Sb.) asiaticus Morawitz asiatica Morawitz in Fedtschenko, 1875:4, examined longiceps Smith, 1878:8 Regeli Morawitz, 1880:337, examined regelii Dalla Torre, 1896:544, unjustified emendation [miniatocaudatus Vogt, 1909:50, infrasubspecific] miniatocaudatus Vogt, 1911:61, examined, not of Vogt,
1909:56 (= B. soroeensis (Fabricius))
heicens Wang, 1982:430, examined, new synonym huangcens Wang, 1982:430, examined, new synonym flavicollis Wang, 1985:163, examined, new synonym baichengensis Wang, 1985:164, examined, new synonym
@ TAXONOMIC STATUS. Several of these nominal taxa have been treated as separate species.
B. heicens, B. huangcens, B. flavicollis and B. baichengensis are morphologically closely similar to B. .asiaticus and differ onlyin details of the colour pattern. In the case of the yellow unbanded colour form and the grey banded colour form in Kashmir (Fig. 12), there is evidence of interbreeding, with many recombinant individuals in some localities (Williams, 1991).
Aside from differences in colour pattern, these taxa are similar in morphology with a range of variation (Williams, 1991). Until more evidence to the contrary is available from critical studies of patterns of varia- tion, I shall treat them as parts of a single variable species.
Bombus (Sb.) niveatus Kriechbaumer niveatus Kriechbaumer, 1870:158 ?vorticosus Gerstaecker, 1872:290, examined (provisional synonym)
@ TAXONOMIC STATUS. B. niveatus and B. vorti- cosus have been regarded both as conspecific (Schmiedeknecht, 1883; Handlirsch, 1888; DallaTorre, 1896; Schulz, 1906) and, more recently, as separate species (e.g. Skorikov, 1922a; Pittioni, 1938; Tkalcii, 1969; Reinig, 1981; Rasmont, 1983).
As far as I am aware, the white-banded B. niveatus occurs only within the broader distributional bounds of the yellow-banded B. vorticosus (within its “extent of occurrence’ in the sense of Gaston, 1994). Although they differ in the colour of the pale pubescence (Pittioni, 1939a), they are closely similar in morphology (Williams, 1991; Baker, 19965). Pittioni (1938) and Baker (1996b) report that they occur at different alti- tudes, without intermediate colour forms. However, the significance of this is unclear, because Baker (1996b) notes that the white-banded B. niveatus co- occurs with other bumble bees (B apollineus (= B. cullumanus), B. simulatilis (= B. ruderarius)) that also show strong convergences in these areas towards the white-banded colour pattern, while elsewhere they are more broadly distributed in yellow-banded colour forms. By analogy with other species (cf. comments
135
on B. melanopygus, B. keriensis), the difference in colour could be the effect of a single pair of alleles for pigment. It is suspicious that both colour forms show identical variation in the extent of pale fringes to the pubescence on the posterior of tergum ILI.
Until more evidence for differences between these nominal taxa other than colour is available from criti- cal studies of patterns of variation, I shall treat them as parts of a single variable species.
Subgenus FRATERNOBOMBUS Skorikov Alpigenobombus (Fraternobombus) Skorikov, 1922a:156, type-species Apathus fraternus Smith (= Bombus fraternus (Smith)) by subsequent designation of Frison, 1927:63 Bombus (Fraternobombus) Franklin, 1954:44
Bombus (Fr.) fraternus (Smith) fraternus (Smith, 1854:385 [Apathus]) examined
136
Subgenus ROBUSTOBOMBUS Skorikov
Volucellobombus Skorikov, 1922a:149, type-species Bombus volucelloides Gribodo (?= B. melaleucus Handlirsch) by monotypy
Alpigenobombus (Robustobombus) Skorikov, 1922a:157, type-species Bombus robustus Smith by subsequent designation of Sandhouse, 1943:597
Bombus (Robustobombus) Richards, 1968:217
COMMENT. Variation within and among the species of this subgenus is particularly poorly understood and a critical review is urgently needed.
Bombus (Rb.) melaleucus Handlirsch melaleucus Handlirsch, 1888:228, examined ?volucelloides Gribodo, 1892:119 (provisional synonym) ?vogti Friese, 1903:254 (provisional synonym) ?nigrothoracicus Friese, 1904:188, examined (provisional synonym) melanoleucus Schulz, 1906:267, unjustified emendation
@ TAXONOMIC STATUS. Several of these nominal taxa have been treated as separate species.
B. volucelloides is closely similar to B. melaleucus, but has been considered to be a separate species (e.g. Milliron, 19735). B. vogti is also closely similar to B. volucelloides, and these two taxa have been consid- ered both as conspecific (e.g. Franklin, 1913; Labougle, 1990) and as separate species (e.g. Milliron, 1973). G. Chavarria (pers. com.) also believes that B. melaleucus, B. volucelloides and B. vogti are all conspecific.
In addition, it seems to me that B. nigrothoracicus is more likely to be conspecific with B. vogti than with B. ecuadorius (see the comments on B. ecuadorius).
Thus B. melaleucus is interpreted here in a very broad sense, to include much variation that is not yet well understood. Until more evidence to the contrary is available from critical studies of patterns of varia- tion, I shall treat them as parts of a single variable species.
P.H. WILLIAMS
O NOMENCLATURE. For this species, the oldest available name of which I am aware is B. melaleucus, which becomes the valid name. The name B. volucelloides has been in most common use, although for just part of this species. However, it seems prema- ture to conserve B. volucelloides by suppressing B. melaleucus until the taxa are better understood, be- cause the name B. melaleucus might yet be required for a separate species or subspecies.
Bombus (Rb.) ecuadorius Meunier Ecuadorius Meunier, 1890:66 ?butteli Friese, 1903:254, examined (provisional syno- nym)
@ TAXONOMIC STATUS. B. butteli is closely simi- lar to B. ecuadorius. They have been considered to be separate species (e.g. Franklin, 1913; Milliron, 1973b), although Franklin conceded that B. butteli (which has grey hairs intermixed on the thoracic dorsum) might be ‘only a variety or subspecies’ of B. ecuadorius (which has the thoracic dorsum en- tirely black).
B. ecuadorius females are very rare in collections. For example, Milliron (19736) had seen only five putative specimens (as opposed to 42 specimens of B. butteli). Of these five specimens, four were fe- males, and just one was a male, which is the same specimen as the holotype of B. nigrothoracicus (see the comments on B. melaleucus). This male is la- belled ‘Bolivia / ?Peru’, whereas the rest of Milliron’s B. ecuadorius are from Ecuador, with the exception of one queen from ‘Peru’ (it carries no further locality data). This putative male of B. ecuadorius differs from the females in having yellow hairs intermixed on the front and rear of the thorax. This was not mentioned in the original description of this male (under the name B. nigrothoracicus) by Friese (1904), which Franklin (1913) used subse- quently as the sole basis for associating the male with B. ecuadorius.
Currently I favour another possible interpretation. This views the male holotype of B. nigrothoracicus instead as a semi-melanic male of B. melaleucus (the males of B. volucelloides [= B. melaleucus] that I have seen have the thoracic dorsum extensively yellow). This might explain the difference in colour pattern and distribution of this male from other B. ecuadorius. However, a consequence of this inter- pretation would be that the only remaining known difference between B. ecuadorius and B. butteli would be in colour pattern, because the main mor- phological justification for regarding them as separate species (the broader apical process of the gonostylus of the putative male B. ecuadorius, now B. melaleucus in the broad sense) would have been removed. Further evidence is awaited.
CHECKLIST OF BUMBLE BEES
Bombus (Rb.) robustus Smith robustus Smith, 1854:400, examined
? Bombus (Rb.) hortulanus Friese
hortulanus Friese, 1904:188, examined
[{hortulans Frison, 1925a:155, incorrect subsequent spell-
ing]
@ TAXONOMIC STATUS. B. robustus and B. hortu- lanus have been considered both as conspecific (e.g. Franklin, 1913; Frison, 1925a; G. Chavarria, pers. com.) and as separate species (e.g. Milliron, 1973; Asperen de Boer, pers.com.).
B. robustus and B. hortulanus are morphologically similar. Among the specimens I have seen, individu- als that have the sides of gastral terga I-II yellow (B. robustus) also have pubescence extending to the mid- dle or almost to the middle of tergum I, and the males have the space between the inner basal process of the gonostylus and the inner apical process narrower than the apical process. Conversely, individuals with the sides of terga I-II black (B. hortulanus) have at least the medial third of tergum I hairless, and the space between the inner processes of the male gono- stylus is wider than the breadth of the apical process.
Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as separate species.
Bombus (Rb.) tucumanus Vachal tucumanus Vachal, 1904:10
Bombus (Sp.) morrisoni Cresson Morrisoni Cresson, 1878:183
Bombus (Sp.) griseocollis (DeGeer) grifeo-collis (DeGeer, 1773:576 [Apis]) separatus Cresson, 1863:165
Bombus (Fn.) funebris Smith funebris Smith, 1854:400, examined
Bombus (Fn.) rohweri (Frison) rohweri (Frison, 1925a:144 [Bremus])
@ TAXONOMIC sTATUS. B. funebris and B. rohweri have been regarded both as conspecific (Milliron, 1962) and as separate species (Frison, 1925a; Asperen de Boer, 1993a; G. Chavarria, pers. com.). They have been distinguished with reference to subtle morpho- logical characters as well as to the consistently and strongly differing colour patterns. Both Asperen de Boer (1993a) and G. Chavarria (pers. com.) found that they co-occur at some localities without intermediate colour patterns. Further evidence is awaited.
Subgenus BRACHYCEPHALIBOMBUS Williams Bombus (Brachycephalibombus) Williams, 1985b:247, type-species Bombus brachycephalus Handlirsch by original designation
138
@ TAXONOMICSTATUS. B. brachycephalus was not explicitly placed in any subgenus by Richards (1968). I described a separate subgenus Brachycephalibombus for B. brachycephalus and B. haueri (Williams, 19855), in order to maintain monophyletic groups (Williams, 1995).
Bombus (Br.) brachycephalus Handlirsch brachycephalus Handlirsch, 1888:244 neotropicus (Frison, 1928:151 [Bremus]) krusemani Asperen de Boer, 1990:1, examined, new syno- nym
@ TAXONOMIC STATUS. The description of B. krusemani shows that this nominal taxon, known from a single location, diverges slightly in colour pattern from the otherwise widespread, common and variable Central American species, B. brachycephalus. The information available at present for B. krusemani is consistent with the known range of variation within B. brachycephalus (e.g. Labougle, 1990). Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (Br.) haueri Handlirsch Haueri Handlirsch, 1888:234
COMMENT. Franklin (1913) and Labougle (1990) believed that this species is closely related to B. crotchii (although Labougle had not examined any males). Surprisingly, Milliron (1973b) placed B. haueri in his ‘Dentatus-group’, without explanation (B. dentatus is a junior synonym of the Indo-Chinese B. breviceps of the subgenus Alpigenobombus). Possibly Milliron, at least, may have been influenced by Skorikoy (1922a), who placed B. haueri in the subgenus Alpigenobombus (as Alpigenobombus (Alpigenobombus) haueri, which he also listed next toAg. (Ag.) crotchii). However, both sexes of species of the subgenus Alpigenobombus, as it has been accepted recently (Richards, 1968; Williams, 1991), are easily distinguished from any New World bumble bees because they have more teeth on the mandibles.
I have examined the morphology of both sexes and, on the basis of cladistic analysis, have grouped B. haueri withB. brachycephalus (Williams, 1985b, 1995) and with B. rubicundus (Williams, 1995). Further evidence is awaited.
Subgenus RUBICUNDOBOMBUS Skorikov Fervidobombus (Rubicundobombus ) Skorikoy, 1922a:154, type-species Bombus rubicundus Smith by subsequent designation of Sandhouse, 1943:597 Bombus (Rubicundobombus) Richards, 1968:217
Bombus (Rc.) rubicundus Smith [Napensis Spinola in Osculati, 1850:201, published with- out description] rubicundus Smith, 1854:400, examined
Bombus (Cc.) coccineus Friese coccineus Friese, 1903:254, examined
Bombus (Cc.) baeri Vachal Baeri Vachal, 1904:10
CHECKLIST OF BUMBLE BEES
@ TAxonomic status. B. handlirschi was not ex- plicitly placed in any subgenus by Richards (1968), and B. macgregori had yet to be described. I have grouped B. handlirschi with B. macgregori in the subgenus Dasybombus (Williams, 1995).
139
Bombus (Ds.) macgregori Labougle & Ayala macgregori Labougle & Ayala, 1985:50, examined menchuae Asperen de Boer, 1995:47, examined, new syno-
nym
@ TAXONOMIC STATUS. B. menchuae was de- scribed from a single location and, on the basis of the worker and male I have examined, appears to diverge from B. macgregori only in colour pattern. Until more evidence to the contrary is available from critical studies of patterns of variation, I shall treat them as parts of a single variable species.
Bombus (Ds.) handlirschi Friese handlirschi Friese, 1903:255, examined
COMMENT. Franklin (1913) believed that this spe- cies is closely related to B. rubicundus. Milliron (1973b) knew ‘of no closely related species in the Western Hemisphere’. I have examined the morphol- ogy of both sexes and, on the basis of cladistic analysis, have grouped B. handlirschi with B. macgregori as sister species (Williams, 1995).
140
REFERENCES
Acerbi, J. 1802. Travels through Sweden, Finland and Lapland, to the North Cape, in the years 1798 and 1799. 2: 376 pp. London. Ackery, P. R. & Vane-Wright, R. I. 1984. Milkweed butterflies, their cladistics and biology, being an account of the natural history of the Danainae, a subfamily of the Lepidoptera, Nymphalidae. 1x+425
pp. London.
Akre, R. D., Greene, A., MacDonald, J. F., Landolt, P. J. & Davis, H. G. 1980. Yellowjackets of America north of Mexico. USDA Handbook No. 552. 102 pp. Washington D.C.
Alford, D. V. 1975. Bumblebees. xii+352 pp. London.
(ed.) 1980. Atlas of the bumblebees of the British Isles. 32 pp. Cambridge.
Amiet, F. 1996. Hymenoptera Apidae, 1. Teil. Allgemeiner Teil, Gattungsschliissel, die Gattungen Apis, Bombus und Psithyrus. Insecta Helvetica (Fauna) 12: 98 pp.
Arretz, P. V. & Macfarlane, R. P. 1982. The introduction of Bombus ruderatus to Chile for red clover pollination. Bee World 67: 15-22.
Ashmead, W. H. 1905. Additions to the recorded Hymenopterous fauna of the Philippine Islands, with descriptions of new species. Proceedings of the United States National Museum 28: 957-971.
Asperen de Boer, J. R. J. van 1990. Bombus krusemani — a new bumblebee species from Guatemala (Hymenoptera: Apidae). Entomologische Berichten 50: \-3.
1992a. Bombus villarricaensis, a new garden bumblebee from
southern Chile (Hymenoptera: Apidae). Entomologische Berichten
52: 133-136.
1992b. Bombus xelajuensis — a new species of bumblebee from Guatemala (Hymenoptera: Apidae). Entomologische Berichten 52: 162-164.
— 1993a. A note onBombus rohweri with a description of the queen (Hymenoptera: Apidae). Entomologische Berichten 53: 32-34. 1993b. Bombus villarricaensis is but a junior synonym of
Bombus ruderatus. Entomologische Berichten 53: 38.
1995. Bombus menchuae — a second species of the subgenus Dasybombus from highland Guatemala (Hymenoptera: Apidae). Entomologische Berichten 55: 47-50.
Baker, D. B. 1996a. On a collection of Bombus and Psithyrus principally from Sutherland, with notes on the nomenclature or status of three species (Hymenoptera, Apoidea). British Journal of Entomology and Natural History 9: 7-19.
— 1996b. On a collection of humble-bees from northern Iran (Hymenoptera: Apoidea, Bombinae). Beitrdge zur Entomologie 46: 109-132.
—— 1996c. The dates of Schmiedeknecht’s Apidae Europaeae. Archives of Natural History 23: 295-298.
1996d. Priorities of publication for some nineteenth-century works describing new taxa of Hymenoptera, principally Apoidea. Archives of Natural History 23: 299-302.
Ball, F. J. 1914. Les bourdons de la Belgique. Annales de la Société entomologique de Belgique 58: 77-108.
Bergstrom, G. & Svensson, B. G. 1973. Studies on natural odorifer- ous compounds, VIII. Characteristic marking secretions of the forms lapponicus and scandinavicus of Bombus lapponicus Fabr. (Hymenoptera, Apidae). Chemica scripta 4: 231-238.
Bingham, C. T, 1897. The fauna of British India, including Ceylon and Burma. Hymenoptera. Vol. I. Wasps and bees. xxix+579 pp. London.
Bischoff, H. 1930. Entomologische Ergebnisse der schwedischen Kamtschatka Expedition 1920-1922. 29. Bombinae (Hymen.). Arkiv for zoologi 21: \-6.
1936. Schwedisch-chinesische wissenschaftliche Expedition
nach den nordwestlichen Provinzen Chinas, unter Leitung von Dr.
Sven Hedin und Prof. Sii Ping-chang. Insekten gesammelt vom
schwedischen Arzt der Expedition Dr. David Hummel 1927-1930.
56. Hymenoptera. 10. Bombinae. Arkiv for zoologi 27: 1-27.
1937. Hymenoptera Aculeata (excl. Formicidae und Halictinae)
von den kanarischen Inseln. Commentationes biologicae 6: 1-3.
P.H. WILLIAMS
Bischoff, H. & Hedicke, H. 1931. Uber einige von Illiger beschriebene Apiden (Hym.). Sitzungsberichte der Gesellschaft naturforschender Freunde zu Berlin 1930: 385-392.
Bright, P. M. & Leeds, H. A. 1938. A monograph of the British aberrations of the chalk-hill blue butterfly Lysandra coridon (Poda, 1761). 144 pp. Bournemouth.
Brown, J. H. 1984. On the relationship between abundance and distribution of species. American Naturalist 124: 255-279.
Burks, B. D. 1951. Tribe Bombini. Jn C. F. W. Muesebeck, K. V. Krombein & H. K. Townes (eds.): Hymenoptera of America north of Mexico. Synoptic catalog. Agricultural Monographs 2: 1247-1255.
Cameron, S. A. & Whitfield, J. B. 1996. Use of walking trails by bees. Nature, London 379: 125.
Cardale, J. C. 1993. Zoological catalogue of Austalia, volume 10. Hymenoptera: Apoidea. ix+406 pp. Canberra.
Castro, L. 1987. Nuevas citas de tres Bombinae (Hym. Apidae) de la Peninstla Ibérica. Boletin de la Asociacién Espanola de Entomologia 11: 413.
1988. Sobre Bombus (Megabombus) reinigiellus (Rasmont,
1983) (Hym., Apidae). Boletin de la Asociacién Espanola de
Entomologia 12: 281-289.
1993. Sobre los Bombinae (Hymenoptera, Apidae) de la Sierra de Albarracin (1). ZAPATERI: revista aragonesa de entomologia 3: 5-13.
Chandler, L. 1950. The Bombidae of Indiana. Proceedings of the Indiana Academy of Science 60: 167-177.
Chiu, S. C. 1948. Revisional notes on the Formosan bombid-fauna (Hymenoptera). Notes d’entomologie chinoise 12: 57-81.
Christ, J. L. 1791. Naturgeschichte, Klassification, und Nomenclatur der Insekten vom Bienen, Wespen, und Ameisengeschlecht; als der fiinften Klasse fiinfte Ordnung des Linneischen Natursystems von dem Insekten: Hymenoptera. Mit hdutigen Fliigeln. 576 pp. Frank- furt am Main.
Claridge, M. F., Dawah, H. A. & Wilson, M. R. 1997. Species: the units of biodiversity. xvi+439 pp. London.
Cockerell, T. D. A. 1905. Descriptions and records of bees. — I. Annals and Magazine of Natural History (7) 16: 216-225.
1906. Descriptions and records of bees. — XII. Annals and
Magazine of Natural History (7) 18: 69-75.
1909. Descriptions and records of bees. — XXIII. Annals and
Magazine of Natural History (8) 4: 393-404.
1910a. Descriptions and records of bees. — XXVIII. Annals and
Magazine of Natural History (8) 5: 409-419.
1910b. Some bees from high altitudes in the Himalaya moun-
tains. Entomologist 43: 238-239.
1911. The humble-bees of Formosa. Entomologist 44: 100-102.
1912. Descriptions and records of bees. — XLV. Annals and
Magazine of Natural History (8) 10: 21-31.
1917. New social bees. Psyche 24: 120-128.
1920. Supplementary notes on the social bees of the Philippine
Islands. Philippine Journal of Science 16: 631-632.
1922. Bees in the collection of the U.S. National Museum. Proceedings of the United States National Museum 60: 20 pp. Colwell, R. K. & Coddington, J. A. 1994. Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of
the Royal Society of London (B) 345: 101-118.
Connor, E. F. & McCoy, E. D. 1979. The statistics and biology of the species-area relationship. American Naturalist 113: 791-833.
Cowan, C. F. 1971. On Guérin’s Iconographie: particularly the insects. Journal of the Society for the Bibliography of Natural History 6: 18-29.
Cracraft, J. 1989. Speciation and its ontology: the empirical conse- quences of alternative species concepts for understanding patterns and processes of differentiation. /n D. Otte & J. A. Endler: Speciation and its consequences, pp. 28-59. Sunderland.
Cresson, E. T. 1863. List of the North American species of Bombus and Apathus. Proceedings of the Entomological Society of Phila- delphia 2: 83-116.
1864. Descriptions of several new species of North American
CHECKLIST OF BUMBLE BEES
Apidae. Proceedings of the Entomological Society of Philadelphia
3: 38-43.
1872. Hymenoptera Texana. Transactions of the American
Entomological Society 4: 153-292.
1874. Descriptions of new Hymenoptera. Transactions of the
American Entomological Society 5: 99-102.
1878. Descriptions of new species of North American bees. Proceedings of the Academy of natural Sciences of Philadelphia 1878: 181-221.
Curtis, J. 1835. Insects. Descriptions, &c. of the insects brought home by Commander James Clark Ross, R.N., ER.S. &c. In J. Ross: Appendix to the narrative of a second voyage in search of a North-West Passage, and of a residence in the arctic regions during the years 1829, 1830, 1831, 1832, 1833, pp. lix-Ixxx. London.
Curtis, J., Haliday, A. H. & Walker, F. 1837. XV. Descriptions, &c. of the insects collected by captain P. P. King, R.N., F.R.S., in the survey of the Straits of Magellan. Transactions of the Linnean Society of London 17: 315-359.
Dahlbom, G. 1832. Bombi Scandinavie monographice tractati et ionibus illustrati. 55 pp. London.
Dalla Torre, K. W. von 1879. Bermerkungen zur Gattung Bombus Latr. I. Bericht der Naturwissenschaftlich-medizinischen Vereins in Innsbruck 8(1877): 3-21.
1880. Unsere hummel- (Bombus) Arten. Der Naturhistoriker 2:
40-41.
1882. Bermerkungen zur Gattung Bombus Latr., I. Bericht des
Naturwissenschaftlich-medizinischen Vereins in Innsbruck 12: 14—
Sih
1890. Hymenopterologische Notizen. Wiener entomologische
Zeitung 9: 139.
1896. Catalogus hymenopterorum hucusque descriptorum systematicus et synonymicus. Volumen X; Apidae (anthophila). viii+643 pp. Lipsiae.
Darwin, C. 1859. On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. ix+490 pp. London.
Day, M. C. 1979. The species of Hymenoptera described by Linnaeus in the genera Sphex, Chrysis, Vespa, Apis and Mutilla. Biological Journal of the Linnean Society 12: 45-84.
DeGeer, C. 1773. Mémoires pour servir a l'histoire des insectes. 3: Viii+696 pp. Stockholm.
Delmas, R. 1976. Contribution a l'étude de la faune frangaise des Bombinae (Hymenoptera, Apoidea, Bombidae). Annales de la Société entomologique de France (Nouvelle serie) 12: 247-290.
Dobzhansky, T. 1937. Genetics and the origin of species. xvit+364 pp. New York.
Edwards, R. 1980. Social wasps, their biology and control. 398 pp. East Grinstead.
Erlandsson, A. 1979. Bombus canariensis Pérez, 1895 n. stat and Bombus maderensis n. sp. from the Macaronesian Islands. Entomologica scandinavica 10: 187-192.
Esmaili, M. & Rastegar, R. 1974. Identified species of aculeate Hymenoptera of Iran. Journal of the Entomological Society of Iran 2: 43-44 + pp. [41-52].
Estoup, A., Solignac, M., Cornuet, J.-M., Goudet, J. & Scholl, A. 1996. Genetic differentiation of continental and island populations of Bombus terrestris (Hymenoptera: Apidae) in Europe. Molecular Ecology 5: 19-31.
Evans, W. 1901. The pale variety of Bombus smithianus, White, in Scotland. Entomologist’s Monthly Magazine 37: 47.
Eversmann, E. 1846. Hymenopterorum Rossicorum species nove vel parum cognite, descripte et ex parte depicte. Byulleten’ Moskovskogo obshchestva ispytatelel prirody 19: 436-443.
1852. Fauna hymenopterologica Volgo-Uralensis. (Continuatio). Familia anthophilarium seu apidarum. /zvéstiva Moskovskago éntomologicheskago obshchestva 3: 3-137.
Fabricius, J. C. 1775. Systema entomologiae, sistens insectorum classes, ordines, genera, species, adiectis synonymis, locis, descriptionibus, observationibus. 32+832 pp. Flensburgi & Lipsiae.
1777. Genera insectorum eorumque characteres naturales se-
141
cundum numerum, figuram, situm et proportionem omnium partium
oris adiecta mantissa specierum nuper detectarum. xvi+310 pp.
Chiloni.
1781. Species insectorum exhibentes eorum differentias
specificas, synonyma auctorum, loca natalia, metamorphosin
adiectis observationibus, descriptionibus. 1: viii+552 pp. Hamburg.
1787. Mantissa insectorum sistens eorum species nuper detectas
adiectis characteribus genericis, differentiis specificis,
emendationibus, observationibus. 1: xx+348 pp. Hafniae.
1793. Entomologia systematica emendata et aucta. Secundum
classes, ordines, genera, species adjectis synonimis, locis
observationibus, descriptionibus. 2: viiit+519 pp. Hafniae.
1798. Supplementum entomologiae systematicae. {2|}+572 pp.
Hafniae.
1804. Systema Piezatorum secundum ordines, genera, species adiectis synonymis, locis, observationibus, descriptionibus. xiv+439 pp. Brunsvigae.
Faester, K. & Hammer, K. 1970. Systematik der mittel- und nordeuropdischen Bombus und Psithyrus (Hym., Apidae). Entomologiske Meddelelser 38: 257-302.
Fisher, R. M. 1987. Queen-worker conflict and social parasitism in bumble bees (Hymenoptera: Apidae). Animal behaviour 35: 1026— 1036.
Forbes, H. O. 1885. A naturalist’s wanderings in the eastern archi- pelago, a narrative of travel and exploration from 1878 to 1883. xix+536 pp. London.
Fourcroy, A. F. de 1785. Entomologia Parisiensis; sive catalogus insectorum que in agro Parisiensi reperiuntur; secundum
methodum Geoffreanam in sectiones, genera & species distributus: cui addita sunt nomina trivialia & fere trecente nove species. 2: 233-544. Paris.
Franklin, H. J. 1907. Notes on Bombinae, with descriptions of new species. Entomological News 18: 90-93.
1911. New North American Bombidae. Transactions of the American Entomological Society 37: 157-168.
— 1913. The Bombidae of the New World. Transactions of the American Entomological Society 38(1912): 177-486.
1915. Notes on Bombidae, with descriptions of new forms
(Hym.). Entomological News 26: 409-417.
1954. The evolution and distribution of American bumblebee kinds. Transactions of the American Entomological Society 80: 43-51.
Friese, H. 1902. Die arktischen Hymenoptera mit Ausschluss der Tenthrediniden. /n F. Romer & F. Schaudinn (eds): Fauna Arctica: eine Zusammenstellung der arktischen Tierformen, mit besonderer Beriicksichtigung des Spitzbergen-Gebietes auf Grund der Ergebnisse der Deutschen Expedition in das Nérdlichen Eismeer im Jahre 1889. 2: 439-SO00. Jena.
1903. Neue Bombus-Arten aus der Neotropischen Region. Zeitschrift fiir Systematische Hymenopterologie und Dipterologie 3: 253-255.
—— 1904. Beitrage zur Bienenfauna von Chile, Peru und Ecuador. Zeitschrift fiir Systematische Hymenopterologie und Dipterologie 4: 180-188.
1905. Neue oder wenig bekannte Hummeln des russischen
Reiches (Hymenoptera). Ezhegodnik Zoologicheskago muzeya
9(1904): 507-523.
1909. Neue Varietaéten von Bombus (Hym.). Deutsche
entomologische Zeitschrift 1909: 673-676.
1910. Neue Bienenarten aus Japan. Verhandlungen des
Zoologisch-Botanischen Gesellschaft in Wien 60: 404-410.
1911. Neue Varietaten von Bombus, Il. Deutsche entomologische Zeitschrift 1911: 571-572.
—— 1912. Zur Synonymie der Bombus-Arten. Deutsche entomologische Zeitschrift 1911: 684.
—— 1913. Uber einige neue Apiden (Hym.). Archiv fiir Naturgeschichte 78(1912)(A,12): 85-89.
1916. Uber einige neue Hummelformen (Bombus), besonders
aus Asien (Hym.). Deutsche entomologische Zeitschrift 1916: 107—
110.
142
1918. Uber Hummelformen aus dem Himalaja. Deutsche
entomologische Zeitschrift 1918: 81-86.
1931. Uber Bombus und Psithyrus. Konowia 10: 300-304.
Friese, H. & Wagner, H. von 1910. Zoologische Studien an Hummeln. I. Die Hummeln der deutschen Fauna. Zoologische Jahrbucher, Systematik, Okologie und Geographie der Tiere 29(1909): 1-104.
Frison, T. H. 1921. New distribution records for North American Bremidae, with description of a new species (Hym.). Entomologi- cal News 32: 144-148.
1923. Systematic and biological notes on bumblebees (Bremidae;
Hymenoptera). Transactions of the American Entomological Soci-
ety 48: 307-326.
1925a. Contribution to the classification of the Bremidae (bum- ble-bees) of Central and South America. Transactions of the American Entomological Society 51: 137-165.
— 1925b. The bumble bees of the Philippine Islands (Bremidae: Hymenoptera). Philippine Journal of Science 27: 113-121.
1927. A contribution to our knowledge of the relationships of
the Bremidae of America north of Mexico (Hymenoptera). Trans-
actions of the American Entomological Society 53: 51—78.
1928. A new species of bumblebee from Guatemala (Bremidae: Hym.). Bulletin of the Brooklyn Entomological Society 23: 151— 152%
— 1930. The bumblebees of Java, Sumatra and Borneo (Bremidae: Hymenoptera). Treubia 12: 1-22.
1933. Records and descriptions of Bremus and Psithyrus from
India (Bremidae: Hymenoptera). Record of the Indian Museum 35:
331-342.
1934. Records and descriptions of Bremus and Psithyrus from Formosa and the asiatic mainland. Transactions of the Natural History Society of Formosa 24: 150-185.
— 1935. Records, notes and descriptions of Bremus from Asia (Bremidae: Hymenoptera). Record of the Indian Museum 37: 339— 363.
Frost, D. R. & Kluge, A. G. 1994. A consideration of epistemology in systematic biology, with special reference to species. Cladistics 10: 259-294.
Gaston, K. J. 1994. Rarity. x+205 pp. London.
Gaston, K. J. & Mound, L. A. 1993. Taxonomy, hypothesis testing and the biodiversity crisis. Proceedings of the Royal Society of London (B) 251: 139-142.
Gaston, K. J., Blackburn, T. M. & Loder, N. 1995. Which species are described first?: the case of North American butterflies. Biodiversity and Conservation 4: 119-127.
Gaston, K. J., Scoble, M. J. & Crook, A. 1995. Patterns in species description: a case study using Geometridae (Lepidoptera). Bio- logical Journal of the Linnean Society 55: 225-237.
Germar, E. F. 1817. Reise nach Dalmatien und in das Gebiet von Ragusa. xiit+323 pp. Leipzig.
Gerstaecker, A. 1869. Beitrage zur naheren Kenntniss einiger Bienen- Gattungen. Stettiner entomologische Zeitung 30: 315-367.
1872. Hymenopterologische Beitrage. Stettiner entomologische Zeitung 33: 250-311.
Ghiselin, M. 1975. A radical solution to the species problem. System- atic Zoology 23(1974): 536-544.
Greene, J. W. 1858. II. Descriptions of several new Hymenopterous insects from the north west coast of America. Annals of the Lyceum of Natural History of New York 7: 11-12.
1860. XXI. Review of the American Bombidae, together with a description of several species heretofore undescribed, being a synopsis of this family of Hymenopterous insects thus far known to inhabit North America. Annals of the Lyceum of Natural History of New York 7: 168-176.
Gribodo, G. 1882. Alcune nuove specie e nuove genere di imenotteri aculeati. Annali del Museo civico di storia naturale Giacomo Doria 18: 261-268.
1892. Contribuzioni imenotterologiche. Sopra alcune specie
nuove 0 poco conosciute di imenotteri antofili (generi Crenoplectra,
Xylocopa, Centris, Psithyrus, Trigona, e Bombus). Bolletino della
Societa Entomologica Italiana 23(1891): 102-119.
P.H. WILLIAMS
Groombridge, B. 1992. Global biodiversity, status of the Earth’s living resources. xx+585 pp. London.
Griitte, E. 1937. Zur Kenntnis zentralasiatischer Arten von Psithyrus Lep. (Hym. Apid.). Mitteilungen der Deutschen entomologischen Gesellschaft 7: 103-109.
Guérin-Méneville, F. E. 1829-1844. Iconographie du régne animal de G. Cuvier, ou représentation d’aprés nature de l'une des espéces les plus remarquables, et souvent non encore figurées, de chaque genre d’animaux. Avec un texte descriptif mis au courant de la science. Ouvrage servir d'atlas a tous les traités de zoologie. Paris.
Gurr, L. 1957. Bumble bee species present in the South Island of New Zealand. New Zealand Journal of Science and Technology (A) 38: 997-1001.
Hagen, H. A. 1862. Bibliotheca entomologica. Die Litteratur tiber das ganze Gebiet der Entomologie bis zum Jahre 1862. 1: xii+512 pp. Leipzig.
Handlirsch, A. 1888. Die Hummelsammlung des k. k. naturhistorischen Hofmuseums. Annalen des Naturhistorischen Museums in Wien 3: 209— 250.
Hanski, I. 1982. Dynamics of regional distribution: the core and satellite species hypothesis. Oikos 38: 210-221.
Harris, M. 1776. An exposition of English insects, with curious observations and remarks, wherein each insect is particularly described; its parts and properties considered; the different sexes distinguished, and the natural history faithfully related. viii+166 pp. London.
Hobbs, G. A. 1966. Ecology of species of Bombus Latr. (Hymenop- tera: Apidae) in southern Alberta. IV. Subgenus Fervidobombus Skorikov. Canadian Entomologist 98: 33-39.
Hoffer, E. 1882. Die Hummeln Steiermarks. Lebensgeschichte und Beschreibung derselben. I. Halfte. [2] + 92 pp. Graz.
1883. Die Hummeln Steiermarks. Lebensgeschichte und
Beschreibung derselben. II. Halfte. 98 pp. Graz.
1889. Die Schmarotzerhummeln Steiermarks. Lebensgeschichte und Beschreibung derselben. 80 pp. Graz.
Hooker, J. D. 1861. Outlines of the distribution of Arctic plants. Transactions of the Linnean Society of London 23: 251-348.
Hoppner, H. 1897. Ueber die bei Freissenbiittel vorkommenden Farbenvarietaéten des Bombus soroensis F. Entomologisches Nachrichtenblatt 23: 329-331.
Horne, C. 1870. Notes on the habits of some hymenopterous insects from the North-west Provinces of India. With an appendix, con- taining descriptions of some new species of Apidae and Vespidae collected by Mr. Horne: by Frederick Smith, of the British Mu- seum. Illustrated by plates from drawings by the author of the notes. Transactions of the Zoological Society of London 7: 161— 196.
Howard, L. O. 1902. The insect book. A popular account of the bees wasps, ants, grasshoppers, flies and other North American insects exclusive of the butterflies moths and beetles, with full life histo- ries, tables and bibliographies. xxvii+429 pp. London.
Hiibner, J. 1816[—1826]. VerzeichniB bekannter Schmettlinge [sic]. 431+72 pp. Augsburg.
Hurd Jr, P. D. 1979. Superfamily Apoidea. Jn K. V. Krombein: Catalog of Hymenoptera in America north of Mexico 2: 1741— 2209. Washington D.C.
Husband, R. W., Fischer, R. L. & Wayne Porter, T. 1980. Descrip- tion and biology of bumblebees (Hymenoptera: Apidae) in Michigan. Great Lakes Entomologist 13: 225-239.
Illiger, J. C. W. 1806. William Kirby’s Familien der bienenartigen Insekten, mit Zusitzen, Nachweisungen und Bemerkungen. Magazin fiir Insektenkunde 5: 28-175.
Imperial Institute of Entomology 1931. Insecta. Zoological Record 68(X1): 1-436.
International Commission on Zoological Nomenclature 1939. Opinion 135. The suppression of the so-called ‘Erlangen List’ of 1801. Opinions and Declarations rendered by the International Commission on Zoological Nomenclature 2: 7-12.
1954. Opinion 220. Validation, under the plenary powers, of the
generic name Bombus Latreille, 1802 (class Insecta, order Hymen-
CHECKLIST OF BUMBLE BEES
optera), in so far as the use of those powers is required to provide
that name with the status of availability. Opinions and Declara-
tions rendered by the International Commission on Zoological
Nomenclature 4: 103-114.
1985. International code of zoological nomenclature. 3rd edn.
xx+338 pp. Berkeley.
1996. Opinion 1828.Apis terrestris Linnaeus, 1758, A. muscorum Linnaeus, 1758 and A. /ucorum Linnaeus, 1761 (currently Bombus terrestris, B. muscorum and B. lucorum) and Bombus humilis Illiger, 1806 (Insecta, Hymenoptera): specific names conserved, and neotypes designated for B. terrestris and B. muscorum. Bulle- tin of Zoological Nomenclature 53: 64-65.
Ito, M. 1985. Supraspecific classification of bumblebees based on the
characters of male genitalia. Contributions from the Institute of
Low Temperature Science, Hokkaido University (B) 20: 143 pp.
— 1993. Chorological note on Japanese bumblebees and the classification of Bombinae (Hymenoptera: Apidae). /n T. Inouye & S. Yamane (eds): Evolution of insect societies, pp. 75-92. Tokyo.
Ito, M. & Munakata, M. 1979. The bumblebees in southern Hokkaido and northernmost Honshu, with notes on Blakiston Zoogeographical Line. Low Temperature Science (B) 37: 81-105.
Ito, M. & Sakagami, S. F. 1980. The bumblebee fauna of the Kurile Islands (Hymenoptera: Apidae). Low Temperature Science (B) 38: 23-51.
Ito, M. & Tadauchi, O. 1981. The occurrence of Psithyrus norvegicus in Kyushu, southern Japan. Kontyii 49: 715.
{Jurine] 1801. Nachricht von einen neuen entomologischen Werke des Hrn. Prof. Jurine in Geneve. /ntelligenzblatt der Literatur- Zeitung 21: 160-165.
Kim, C.-W. & Ito, M. 1987. On the bumblebees from the Korean Peninsula (Hymenoptera, Bombidae). Entomological Research Bul- letin 13: 1-42.
Kirby, W. 1802. Monographia apumAngliae; or, an attempt to divide into their natural genera and families, such species of the Linnean genus Apis as have been discovered in England: with descriptions and observations. 2: 388 pp. Ipswich.
1837. Part 4. The insects. Jn J. Richardson: Fauna boreali- Americana; or the zoology of the northern parts of BritishAmerica: containing descriptions of the objects of natural history collected on the late Northern Land expeditions, under command of Captain Sir John Franklin, R.N. XXxix+325 pp. Norwich.
Klug, F. 1807. Species apiariarum familiae novas, descripsit, generumque characteres adjecit. Magazin Gesellschaft Naturforschender Freunde zu Berlin 1: 263-265.
Kriechbaumer, J. 1870. Vier neue Hummelarten. Verhandlungen der Zoologisch-botanischen Gesellschaft in Wien 20: 157-160.
— 1877. Bombus Mocsaryi n. sp. Stettiner entomologische Zeitung 38: 253-254.
Kriiger, E. 1917. Zur Systematik der mitteleuropaischen Hummeln (Hym.). Entomologische Mitteilungen 6: 55-66.
1920. Beitrage zur Systematik und Morphologie der mittel- europaischen Hummeln. Zoologische Jahrbiicher (Systematik, Okologie und Geographie der Tiere) 42: 289-464.
— 1939. Die Hummeln und Schmarotzerhummeln von Sylt und dem benachbarten Festland. Schriften des Naturwissenschaftlichen Vereins fiir Schleswig-Holstein 23; 28-123.
1951. Phanoanalytische Studien an einigen Arten der
Untergattung Terrestribombus O. Vogt (Hymen. Bomb.). I. Teil.
Tijdschrift voor Entomologie 93(1950): 141-197.
1954. Phaenoanalytische Studien an einigen Arten der Untergattung Terrestribombus O. Vogt (Hymenoptera, Bombidae). Il. Teil. Tijdschrift voor Entomologie 97: 263-298.
— 1956. Phaenoanalytische Studien an einigen Arten der Untergattung Terrestribombus O. Vogt (Hymenoptera, Bombidae). Il. Teil. Tijdschrift voor Entomologie 99: 75-105.
— 1958. Phaenoanalytische Studien an einigen Arten der Untergattung Terrestribombus O. Vogt (Hymenoptera, Bombiidae). Ill. Teil. Tijdschrift voor Entomologie 101: 283-344.
Kruseman, G. 1952. Subgeneric division of the genus Bombus Latr. Transactions of the 9th International Congress of Entomology,
143
Amsterdam 1: 101-103.
LaBerge, W. E. & Webb, M. C. 1962. The bumblebees of Nebraska. Research Bulletin, University of Nebraska College of Agriculture, Agricultural Experiment Station 205: 38 pp.
Labougle, J. M. 1990. Bombus of México and Central America (Hymenoptera, Apidae). Kansas University Science Bulletin 54: 35-73.
Labougle, J. M. & Ayala, R. 1985. A new subgenus and species of Bombus (Hymenoptera: Apidae) from Guerrero, Mexico. Folia Entomolégica Mexicana 66: 47-55.
Labougle, J. M., Ito, M. & Okazawa, T. 1985. The species of the genus Bombus (Hymenoptera: Apidae) of Chiapas, Mexico and Guatemala; with a morphometric and altitudinal analysis. Folia Entomolégica Mexicana 64: 55-72.
Latreille, P. A. 1802a. Histoire naturelle des fourmis, et recueil de mémoires et d’observations sur les abeilles, les araignées, les faucheurs, et autres insectes. xvi+445 pp. Paris.
1802b. Histoire naturelle, générale et particuliére des crustaces
et des insectes. 3: xii+467 pp. Paris.
1809. Genera crustaceorum et insectorum, secundum ordinem
naturalem in familias disposita, iconibus exemplisque plurimis explicata. Tomus quartus et ultimus. 397 pp. Paris.
Laverty, T. M. & Harder, L. D. 1988. The bumble bees of eastern Canada. Canadian Entomologist 120: 965-987.
Laverty, T. M., Plowright, R. C. & Williams, P. H. 1984. Digressobombus, a new subgenus with a description of the male of Bombus digressus (Hymenoptera: Apidae). Canadian Entomolo- gist 116: 1051-1056.
Lepeletier de Saint-Fargeau, A. L. M. 1832. Observations sur louvrage intitulé: ‘bombi scandinaviae monographice tractato, etc., 4 Gustav. Dahlbom.’. Annales de la Société Entomologique de France 1: 366-382.
1836. Histoire naturelle des insectes. Hyménopteéres. 1: 547pp. Paris.
Linnaeus, C. 1758. Systema naturae per regna tria naturae, secun-
dum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. 823 pp. Holmiae.
— 1761. Fauna Suecica sistens animalia Suecica regni: Mamma- lia, Aves, Amphibia, Pisces, Insecta, Vermes. Distributa per classes & ordines, genera & species, cum differentiis specierum, synonymis auctorum, nominibus incolarum, locis natalium, descriptionibus insectorum. [45]+578 pp. Stockholmiae.
1790. Systema naturae per regna tria naturae, secundum classes,
ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tom. I. Pars V. pp. 2225-3020. Lipsiae.
Loken, A. 1973. Studies on Scandinavian bumble bees (Hymenop- tera, Apidae). Norsk entomologisk Tidsskrift 20: 1-218.
1984. Scandinavian species of the genus Psithyrus Lepeletier (Hymenoptera: Apidae). Entomologica scandinavica (supplement) 23: 45 pp.
Loken, A. & Framstad, E. B. 1983. Contribution to the taxonomy of Psithyrus (Fernaldaepsithyrus) (Hymenoptera: Apidae). Acta entomologica fennica 42: 46-SO0.
Loken, A., Pekkarinen, A. & Rasmont, P. 1994. Case 2638. Apis terrestris Linnaeus, 1758, A. muscorum Linnaeus, 1758 and A. lucorum Linnaeus, 1761 (currently Bombus terrestris, B. muscorum and B. lucorum) and Bombus humilis Mliger, 1806 (Insecta, Hy- menoptera): proposed conservation of usage of the specific names. Bulletin of Zoological Nomenclature 51; 232-236.
Maa, T. 1948. On some eastern asiatic species of the genus Psithyrus Lepel. (Hymenoptera: Bombidae). Notes d’entomologie chinoise 12: 17-37.
Macfarlane, R. P. & Gurr, L. 1995. Distribution of bumble bees in New Zealand. New Zealand Entomologist 18: 29-36.
Macfarlane, R. P., Patten, K. D., Royce, L. A., Wyatt, B. K. W. & Mayer, D. F. 1994. Management potential of sixteen North Ameri- can bumble bee species. Melanderia 50: 1-12.
Macior, L. W. 1975. The pollination ecology of Pedicularis (Scrophulariaceae) in the Yukon Territory. American Journal of Botany 62: 1065-1072.
144
— 1990. Pollination ecology of Pedicularis punctata Decne. (Scrophulariaceae) in the Kashmir Himalaya. Plant Species Biol- ogy 5: 215-223.
Mallet, J. 1995. A species definition for the modern synthesis. Trends in Ecology and Evolution 10: 294-299.
— 1997. What are species? Trends in Ecology and Evolution 12: 453-454.
Martin, G. 1996. Birds in double trouble. Nature, London380: 666-667.
Matsumura, S. 1932. Illustrated common insects of Japan. IV. Hy- menoptera, Diptera, Rhyncota. {1|+253 pp. Tokyo.
May, J. 1944. Beitrag zur Systematik einiger Arten der Schmarotzerhummeln — Psithyrus Lep. (Hymen. Apoid.). Sbornik Entomologického oddeléni Narodniho musea v Praze 21-22: 231- DiS:
Mayr, E. 1940. Speciation phenomena in birds. American Naturalist 74: 249-278.
1963. Animal species and evolution. xiv+797 pp. Massachu- setts.
Medler, J. T. & Carney, D. W. 1963. Bumblebees of Wisconsin (Hymenoptera: Apidae). Research Bulletin, University of Wiscon- sin, Agricultural Experiment Station 240: 47 pp.
Menge, F. A. 1856. Lebenszeichen vorweltlicher, im Bernstein eingeschlossener Thiere. 32 pp. Danzig.
Menke, A. S. & Carpenter, J. 1984. Nuclearbombus, new subgenus (or how to eliminate bumblebee subgenera and learn to love Bombus). Sphecos 9: 28.
Meunier, F. 1890. Description d'une espéce nouvelle «ou peu connue» de Bombus d’ Ecuador. Jornal de Sciencias Mathematicas, Physicas e Naturaes (2) 2: 66.
Michener, C. D. 1979. Biogeography of the bees. Annals of the Missouri Botanical Garden 62: 335-353.
1990. Classification of the Apidae (Hymenoptera). University of Kansas Science Bulletin 54: 75-164.
Middendorff, A. T. von 1851. Reise in den Aussersten Norden und Osten Sibiriens, wahrend der Jahre 1843 und 1844 mit allerhéchster Genehmigung auf Veranstaltung der kaiserlichen Akademie der Wissenschaften zu St. Petersburg ausgefiihrt und in Verbindung mit vielen Gelehrten herausgegeben. Band II. Zoologie. Theil 1. 516 pp. St. Petersburg.
Milliron, H. E. 1960. Recognition of bumblebee type specimens, with notes on some dubious names (Hymenoptera: Apidae). Bulle- tin of the Brooklyn Entomological Society 55: 87-99.
1961. Revised classification of the bumblebees — a synopsis
(Hymenoptera: Apidae). Journal of the Kansas Entomological
Society 34: 49-61.
1962. Taxonomic notes on some American bumblebees (Hy-
menoptera: Apidae; Bombinae). Canadian Entomologist 94:
728-735.
1970a. Pyrobombus (P.) cascadensis, an undescribed species of
bumblebee from the Pacific Northwest, U.S.A. (Hymenoptera:
Bombinae). Canadian Entomologist 102: 382-383.
1970b. A monograph of the western hemisphere bumblebees (Hymenoptera: Apidae; Bombinae). Bibliography. Memoirs of the Entomological Society of Canada 65: \ii pp.
—— 1971. A monograph of the western hemisphere bumblebees (Hymenoptera: Apidae; Bombinae). I. The genera Bombus and Megabombus subgenus Bombias. Memoirs of the Entomological Society of Canada 82: 1-80.
—— 1973a. A monograph of the western hemisphere bumblebees (Hymenoptera: Apidae; Bombinae). II]. The genus Megabombus subgenus Megabombus. Memoirs of the Entomological Society of Canada 89: 81-237.
—— 1973b. A monograph of the western hemisphere bumblebees (Hymenoptera: Apidae; Bombinae). III. The genus Pyrobombus subgenus Cullumanobombus. Memoirs of the Entomological Soci- ety of Canada 91: 238-333.
Milliron, H. E. & Oliver, D. R. 1966. Bumblebees from northern Ellesmere Island, with observations on usurpation by Megabombus hyperboreus (Schénh.) (Hymenoptera: Apidae). Canadian Ento- mologist 98: 207-213.
P.H. WILLIAMS
Mitchell, T. B. 1962. Bees of the eastern United States. Volume 2. Technical Bulletin, North Carolina Agricultural Experiment Sta- tion 152: 557 pp.
Morawitz, F. 1869. Die Bienen des Gouvernements von St. Petersburg. Trudy Russkago éntomologicheskago obshchestva 6: 27-71.
—— 1874. Die Bienen Daghestans. Trudy Russkago éntomologicheskago obshchestva 10(1873): 129-189.
1875. [Bees.] (Mellifera). In: A. Fedtschenko: Reise in Turkestan. II Zoologischer Teil, pp. i-ii, 1-303. Berlin.
—— 1880. Ein Beitrag zur Bienen-Fauna mittel-Asiens. lzvéstiya Imperatorskoi akademii nauk 26: 337-379.
1881. Die russischen Bombus-Arten in der Sammlung der
Kaiserlichen Academie der Wissenschaften. /zvéstiva.
Imperatorskoi akademii nauk 27: 213-265.
1883. Neue russisch-asiatische Bombus-Arten. Trudy Russkago
éntomologicheskago obshchestva 17: 235-245.
1886. Insecta in itinere Cl. N. Przewalskii in Asia centrali
novissime lecta. I. Apidae. 7rudy Russkago éntomologicheskago
obshchestva 20: 195-229.
1888. Hymenoptera aculeata nova. Descripsit. Trudy Russkago
éntomologicheskago obshchestva 22: 224-302.
1890. Insecta a Cl. G. N. Potanin in China et in Mongolia
novissime lecta. XIV. Hymenoptera Aculeata. II). II. Apidae.
Trudy Russkago éntomologicheskago obshchestva 24: 349-385.
1892. Hymenoptera aculeata Rossica nova Trudy Russkago
éntomologicheskago obshchestva 26: 132-181.
1894. Supplement zur Bienenfauna Turkestans. 7rudy Russkago éntomologicheskago obshchestva 28: 1-87.
Morice, F. D. & Durrant, J. H. 1915. The authorship and first publication of the ‘Jurinean’ genera of Hymenoptera: being a reprint of a long-lost work by Panzer, with a translation into English, an introduction, and bibliographical and critical notes. Transactions of the Entomological Society of London 47(1914): 339-436.
Moure, J. S. & Sakagami, S. F. 1962. As mamangabas sociais do Brasil (Bombus Latreille) (Hymenoptera, Apoidea). Studia entomologica 5: 65-194.
Miiller, O. F. 1776. Zoologia Danice prodromus, seu animalium Danie et Norvegie indigenarum characteres, nomina, et synonyma imprimis popularium. xxxii+274 pp. Havnie.
Nelson, G. 1972. Phylogenetic relationship and classification. Sys- tematic Zoology 21: 227-231.
Newman, E. 1835. Attempted division of British insects into natural orders. Entomological Magazine 2(1834): 379-431.
Nylander, W. 1848. Adnotationes in expositionem monographicam apum borealium. Meddelanden af Societatis pro fauna et flora fennica 1: 165-282.
Obrecht, E. & Scholl, A. 1981. Enzymelektrophoretische Untersuchungen zur Analyse der Verwandtschaftsgrade zwischen Hummel- und Schmarotzerhummelarten (Apidae, Bombini). Apidologie 12: 257-268.
Oliff, A. S. 1895. Successful introduction of humble bees into New South Wales. Entomologist’s Monthly Magazine 31: 67.
Olivier, G. A. 1789. Encyclopédie méthodique. Histoire naturelle. Tome quatrieme. Insectes. ccclxxili+331 pp. Paris.
Ornosa, C. 1986a. Revisi6n de los Bombinae estudiados por D. Modesto Quilis (Hym., Apidae). Boletin de laAsociacion Espanola de Entomologia 10: 359-368.
1986b. Especies de Bombinae de la Sierra de los Ancares
(Hymenoptera, Apidae). Actas de las VIII Jornadas Asociacion
Espanola de Entomologia pp. 1042-1050.
1991. Consideraciones taxonémicas sobre el subgénero Megabombus Dalla Torre, 1880, en la Peninsula Ibérica (Hym., Apidae, Bombinae). Anales de Biologia 17: 33-41.
Osculati, G. 1850. Esplorazione delle regioni equatoriali lungo il napo ed il fiume delle Amazzoni frammento di un viaggio fatto nelle due Americhe negli anni 1846—1847—1848. 320 pp. Milano.
Owen, R. E. & Plowright, R. C. 1980. Abdominal pile color dimor- phism in the bumble bee, Bombus melanopygus. Journal of Heredity 71: 241-247.
SO
ee eee a ee
CHECKLIST OF BUMBLE BEES
Ozbek, H. 1983. Taxonomical and some biological studies on Bombinae (Hymenoptera: Apoidea, Bombinae) from some parts of eastern Anatolia. Atatiirk Universitesi Yayinlari No. 631, Ziraat Fakiiltesi Yayinlari No. 287, Arastirmalar Serisi No. 188, 70 pp.
1990. A new bumblebee species of Pyrobombus Dalla Torre
(Hymenoptera, Apidae, Bombinae) in eastern Anatolia, Turkey.
Tiirkiye entomoloji dergisi 14: 207-214.
1997. Bumblebees fauna of Turkey with distribution maps (Hymenoptera: Apidae: Bombinae) Part 1: Alpigenobombus Skorikov, Bombias Robertson and Bombus Latreille. Tiirkiye entomoloji dergisi 21: 37-56.
Pagliano, G. 1995. Hymenoptera Apoidea. /n A. Minelli, S. Ruffo & S. La Posta (eds): Checklist delle specie delle fauna Italiana 106(1994): 23-24. Bologna.
Pallas, P. S. 1771. Reise durch verschiedene Provinzen des ruBischen Reichs. 1: 504 pp. St Petersburg.
Pamilo, P., Pekkarinen, A. & Varvio, S.-L. 1987. Clustering of bumblebee subgenera based on interspecific genetic relationships (Hymenoptera, Apidae: Bombus and Psithyrus). Annales zoologici fennici 24: 19-27.
Pamilo, P., Teng6, J., Rasmont, P., Pirhonen, K., Pekkarinen, A. & Kaarnama, E. 1997. Pheromonal and enzyme genetic characteris- tics of the Bombus lucorum species complex in northern Europe. Entomologica fennica 7: 187-194.
Pamilo, P., Varvio-Aho, S.-L. & Pekkarinen, A. 1984. Genetic variation in bumblebees (Bombus, Psithyrus) and putative sibling species of Bombus lucorum. Hereditas 101; 245-251.
Panfiloy, D. V. 1951. Bumblebees of the subgenus Cullumanobombus Vogt (Hymenoptera, Apoidea). Trudy Vsesoyuznogo éntomologicheskago obshchestva 43: 115-128.
1956. Contribution to the taxonomy of bumblebees (Hymenop- tera, Bombinae), including the description of new forms. Zoologicheskti Zhurnal 35: 1325-1334.
Panzer, G. W. F. 1801. Faunae insectorum Germanicae initia oder Deutschlands Insecten gesammelt und herausgegeben. Heft 73-84. Niirnberg.
1805. Faunae insectorum Germanicae initia oder Deutschlands Insecten gesammelt und herausgegeben. Heft 8S—90. Niirnberg. Parry, W. E. 1824. Journal of a voyage for the discovery of a North-
West Passage from the Atlantic to the Pacific; performed in the years 1819-20, in His Majesty's ships Hecla and Griper, under the orders of William Edward Parry, R.N., F-R.S., and commander of the expedition. With an appendix, containing the scientific obser-
vations. Xxix+310+cccx pp. London.
Paterson, H. E. H. 1985. The recognition concept of species. /n E. S. Vrba: Species and speciation, pp. 21-29. Transvaal Museum Mono- graph No. 4. Pretoria.
Patterson, B. D. 1996. The ‘species alias’ problem. Nature, London 380: 589.
Pekkarinen, A. 1979. Morphometric, colour and enzyme variation in bumblebees (Hymenoptera, Apidae, Bombus) in Fennoscandia and Denmark. Acta zoologica fennica 158: 60 pp.
— 1982. Morphology and specific status of Bombus lapponicus (Fabricius) and B. monticola Smith (Hymenoptera: Apidae). Entomologica scandinavica 13: 4146.
Pekkarinen, A. & Kaarnama, BE. 1994. Bombus terrestris auct. new to Finland (Hymenoptera, Apidae). Sah/bergia 1: 11-13.
Pekkarinen, A. & Teriis, I. 1993. Zoogeography of Bombus and Psithyrus in northwestern Europe (Hymenoptera , Apidae). Annales Zoologici Fennici 30: 187-208.
Pekkarinen, A., Teras, I., Viramo, J. & Paatela, J. 1981. Distribu- tion of bumblebees (Hymenoptera, Apidae: Bombus and Psithyrus) in eastern Fennoscandia. Notulae Entomologicae 61: 71-89.
Pendlebury, H. M. 1923. Four new species of Bombus from the Malay Peninsula. Journal of the Federated Malay States Museums 11: 64-67.
Pérez, J. 1880. Contribution 4 la faune des apiaires de France. Premiére partie, récoltantes. Actes de la Société linnéenne de Bordeaux 33(1879): 119-229.
1884. Contribution a la faune des apiaires de France. Deuxiéme
145
partie. Parasites. Actes de la Société linnéenne de Bordeaux 37(1883): 205-380.
— 1890. Catalogue des melliféres du sud-ouest. Actes de la Société linnéenne de Bordeaux 44: 133-200.
— 1895. Voyage de M. Ch. Alluaud aus iles Canaries (Novembre 1889-Juin 1890), 4e mémoire (1). Hymeénopteres. Annales de la Société entomologique de France 64: 191-201.
1905. Hyménopteres recueillis dans le Japon central, par M.
Harmand, ministre plénipotentiaire de France a Tokio. Bulletin du
Muséum national d'histoire naturelle 11: 23-39.
1909. Sur quelques variétés de bourdons de la Corse. Procés- verbaux de la Société linnéenne de Bordeaux 62: clvii—clix.
Pittioni, B. 1937. Die Hummelfauna des Kalsbachtales in Ost-Tirol. Ein Beitrag zur Okologie und Systematik der Hummeln Mitteleuropas. Festschrift zum 60. Geburtstag von Professor Dr. Embrik Strand 3: 64-115.
1938. Die Hummeln und Schmarotzerhummeln der Balkan-
Halbinsel. Mit besonderer Beriicksichtigung der Fauna Bulgariens.
I. Allgemeiner Teil. /zvestiya na Tsarskite prirodonauchni instituti
v Sofiya 11: 12-69.
1939a. Die Hummeln und Schmarotzerhummeln der Balkan-
Halbinsel. Il. Spezieller Teil. /zvestiya na Tsarskite prirodonauchni
instituti v Softya 12: 49-115.
1939b. Bombus (Agrobombus) bureschi sp. nov. (Hymenopt., Apidae), eine neue Hummelart von der Balkanhalbinsel und einige weitere interessante neue Hummelformen. Arbeiten der Bulgarischen Naturforschenden Gesellschaft 18: \-10.
— 1939c. Neue und wenig bekannte Hummeln der Palaarktis (Hymenopt., Apidae). Konowia 17(1938): 244-263.
— 1939d. Tanguticobombus subg. nov. (Hymenopt., Apidae). Zoologischer Anzeiger 126: 201-205.
— 1942. Die boreoalpinen Hummeln und Schmarotzerhummeln (Hymen., Apidae, Bombinae). I. Teil. /zvestiya na Tsarskite prirodonauchni instituti v Sofiya 15: 155-218.
1943. Die boreoalpinen Hummeln und Schmarotzerhummeln
(Hymen., Apidae, Bombinae). Il. Teil. /zvestiva na Tsarskite
prirodonauchni instituti v Sofiya 16: \-77.
1949. Beitrage zur Kenntnis der Bienenfauna SO-Chinas. Die Hummeln und Schmarotzerhummeln der Ausbeute J. Klapperich (1937/38). (Hym., Apoidea, Bombini). Eos 25: 241-284.
Plowright, R. C. & Owen, R. E. 1980. The evolutionary significance of bumble bee color patterns: a mimetic interpretation. Evolution 34: 622-637.
Plowright, R. C. & Stephen, W. P. 1973. A numerical taxonomic analysis of the evolutionary relationships of Bombus and Psithyrus (Apidae: Hymenoptera). Canadian Entomologist 105: 733-743.
— 1980. The taxonomic status of Bombus franklini (Hymenoptera: Apidae). Canadian Entomologist 112: 475-479.
Poole, R. W. 1996. Nomina insecta nearctica [sic], a checklist of the insects of North America. Volume 2: Hymenoptera, Mecoptera, Megaloptera, Neuroptera, Rhaphidioptera, Trichoptera. 793 pp. Rockville, Maryland.
Popoy, V. B. 1927. New forms of the genus Psithyrus Lep. Konowia 6; 267-274.
1930. Note on Agrobombus smithianus White (Hymenoptera,
Bombidae). Russkoe éntomologicheskoe obozrénie 24: 95-99.
1931. Zur Kenntnis der palaarktischen Schmarotzerhummeln (Psithyrus Lep.). Eos 7: 131-209.
Prys-Jones, O. E. & Corbet, S. A. 1987. Bumblebees. 86 pp. Cambridge.
Prys-Jones, O. E., Olafsson, E. & Kristjansson, K. 1981. The Icelandic bumble bee fauna (Bombus Latr., Apidae) and its dis- tributional ecology. Journal of Apicultural Research 20: 189-197.
Queiroz, K. de & Donoghue, M. J. 1988. Phylogenetic systematics and the species problem. Cladistics 4: 317-338.
Quilis-Pérez, M. 1927. Los apidos de Espafia género Bombus Latr. Anales del Instituto Nacionale, Valencia 15: 1-119.
Radoszkowski, O. 1859. Sur quelques hyménopteres nouveaux ou peu connus de la collection du Musée de 1’ Académie des sciences
146
de St. Pétersbourg. Byulleten’Moskovskogo obshchestva ispytatelet prirody 32: 479-486.
— 1862. Sur quelques hyménoptéres nouveaux ou peu connus. Byulleten’ Moskovskogo obshchestva ispytatelet prirody 35: 589- 598.
—— 1876. Matériaux pour servir 4 une faune hyménoptérologique de la Russie. Zrudy Russkago éntomologicheskago obshchestva 12: 82-110.
1877a. In: Séance du 3 (15) Mai 1876. Trudy Russkago
éntomologicheskago obshchestva 13: vii-viii.
1877b. Essai d'une nouvelle méthode pour faciliter la
détermination des espéces appartenant au genre Bombus. Byulleten’
Moskovskogo obshchestva ispytatelei prirody 52: 169-219.
1881. Jn: Sitzung am 5 (17) Februar. 7rudy Russkago
éntomologicheskago obshchestva 16: v—vii.
1883. Sur quelques espéces Russes appartenant au genre Bombus.
Byulleten’ Moskovskogo obshchestva ispytatelel prirody 58: 168—
226.
1884. Révision des armures copulatrices des males du genre
Bombus. Byulletin’ Moskovskogo obshchestva ispytatelel prirody
59: 51-92.
1888. Etudes hyménopterologiques. I. Révision des armures
copulatrices des males. 1) Bombus. Trudy Russkago
éntomologicheskago obshchestva 22: 315-337.
1893. Descriptions d’hyménoptéres nouveaux. Revue d’entomologie 12: 241-245.
Rasmont, P. 1982. A propos des bourdons (Hymenoptera, Apidae) de la Corse. Bulletin de la Société entomologique de Mulhouse 1982: 49-61.
1983. Catalogue commenté des bourdons de la région ouest-
paléarctique (Hymenoptera, Apoidea, Apidae). Notes Fauniques de
Gembloux 7: 71 pp.
1984. Les bourdons du genre Bombus Latreille sensu stricto en
Europe occidentale et centrale (Hymenoptera, Apidae). Spixiana 7:
135-160.
1988. Monographie écologique et zoogéographique des bourdons de France et de Belgique (Hymenoptera, Apidae, Bombinae). 309+1xi pp. Ph.D. thesis, Faculté des Sciences agronomiques de |’Etat, Gembloux.
Rasmont, P. & Adamski, A. 1995. Les bourdons de la Corse (Hy- menoptera, Apoidea, Bombinae). Notes Fauniques de Gembloux 31: 3-87.
Rasmont, P., Delmas, R., Leclant, F. & Yarrow, I. H. H. 1987. Megabombus (Rhodobombus) mesomelas (Gerstaecker, 1869) (Hy- menoptera, Apidae, Bombinae). Documents pour un Atlas zoogéographique du Languedoc-Roussillon 32: 4 pp.
Rasmont, P., Ebmer, P. A., Banaszak, J., & Zanden, G. van der 1995. Hymenoptera Apoidea Gallica. Liste taxonomique des abeilles de France, de Belgique, de Suisse et du Grand-Duché de Luxembourg. Bulletin de la Société de France 100: 1-98.
Reinig, W. F. 1930. Untersuchungen zur Kenntnis der Hummelfauna des Pamir-Hochlandes. Zoologische Ergebnisse der deutsch- russischen Alai-Pamir-Expedition der Notgemeinschaft der Deutschen Wissenschaft und der Akademie der Wissenschaften der U.d.S.S.R. Zeitschrift fiir Morphologie und Okologie der Tiere 17: 68-123.
1934. Entomologische Ergebnisse der deutsch-russischen Alai-
Pamir-Expedition, 1928 (III). 7. Hymenoptera VIII (Gen. Bombus
Fabr.). Nachtrag. Deutsche entomologische Zeitschrift 1933: 163—
174.
1935. On the variation of Bombus lapidarius L. and its cuckoo, Psithyrus rupestris Fabr., with notes on mimetic similarity. Journal of Genetics 30: 321-356.
—— 1936. Beitrige zur Kenntnis der Hummelfauna von Mandschukuo (Hym. Apid.). Mitteilungen der Deutschen entomologischen Gesellschaft 7: 2-10.
1939. Die Evolutionsmechanismen, erlautert an den Hummeln.
Verhandlungen der Deutschen zoologischen Gesellschaft (supple-
ment) 12: 170-206.
1940. Beitrage zur Kenntnis der Hummelfauna von Afghani-
P.H. WILLIAMS
stan. (Hym., Apid.). (Ergebnisse der Reise von H. und E. Kotzsch
in den Hindukusch im Jahre 1936.) Deutsche entomologische
Zeitschrift 1940: 224-235.
1970. Bastardierungszonen und Mischpopulationen bei
Hummeln (Bombus) und Schmarotzerhummeln (Psithyrus).
(Hymenopt., Apidae). Mitteilungen der Miinchener
entomologischen Gesellschaft 59(1969): 1-89.
1971. Zur Faunistik und Zoogeographie des Vorderen Orients.
3. Beitrag zur Kenntnis der Hummeln und Schmarotzerhummeln
Anatoliens (Hym., Apidae). Veroffentlichungen der Zoologischen
Staatssammlung 15: 139-165.
1972. Okologische Studien an mittel- und siidosteuropaischen
Hummeln (Bombus Latr., 1802; Hym., Apidae). Mitteilungen der
Miinchner entomologischen Gesellschaft 60(1970): 1-56.
1974. Zur Verbreitung einiger Hummelarten auf der Balkan-
Halbinsel (Hym., Bombidae). Nachrichtenblatt der Bayerischen
Entomologen 23: 11-13.
1981. Synopsis der in Europa nachgewiesenen Hummel- und Schmarotzerhummelarten (Hymenoptera, Bombidae). Spixiana 4: 159-164.
Richards, K. W. 1973. Biology of Bombus polaris Curtis and B. hyperboreus Schonherr at Lake Hazen, Northwest Territories (Hy- menoptera: Bombini). Quaestiones entomologicae 9: 115-157.
Richards, O. W. 1928a. On a collection of humble-bees (Hymenop- tera, Bombidae) made in Ladakh by Col. R. Meinertzhagen. Annals and Magazine of Natural History (10) 2: 333-336.
1928b. A revision of the European bees allied to Psithyrus
quadricolor, Lepeletier (Hymenoptera, Bombidae). Transactions
of the Entomological Society of London 76: 345-365.
1929a. On two new species of humble-bees in the collection of the British Museum, constituting a new group of the genus Psithyrus, Lep. (Hymenoptera, Bombidae). Annals and Magazine of Natural History (10) 3: 139-143.
— 1929b. The types of the humble-bees described by Gribodo (Hymenoptera, Bombidae). The Entomologist’s Monthly Maga- zine 65: 58-59.
— 1929c. A revision of the humble-bees allied toBombus orientalis, Smith, with the description of a new subgenus. Annals and Maga- zine of Natural History (10) 3: 378-386.
1930. The humble-bees captured on the expeditions to Mt.
Everest (Hymenoptera, Bombidae). Annals and Magazine of Natu-
ral History (10) 5: 633-658.
193la. Some notes on the humble-bees allied to Bombus
alpinus, L. Troms¢ Museums Arshefter 50(1927): 1-32.
1931b. A new species of Indian humble-bee in the collection of the British Museum (Hymenoptera, Bombidae). Annals and Maga- zine of Natural History (10) 8: 529-533.
—— 1934. Some new species and varieties of oriental humble-bees (Hym. Bombidae). Stylops 3: 87-90.
— 1935. Bombus muscorum (Linnaeus) and B. smithianus White (Hym.). Transactions of the Society for British Entomology 2: 73-85.
1951. The 3rd Danish expedition to Central Asia. Zoological results 5. Bombidae (Insecta) from Afghanistan. Videnskabelige Meddelelser fra Dansk naturhistorisk Forening i Kjgbenhavn 113: 191-199.
—— 1968. The subgeneric divisions of the genus Bombus Latreille (Hymenoptera: Apidae). Bulletin of the British Museum (Natural History) (Entomology) 22: 209-276.
Robertson, C. 1903. Synopsis of Megachilide and Bombine. Trans- actions of the American Entomological Society 29: 163-178.
Rosen, D. E. 1979. Fishes from the uplands and intermontane basins of Guatemala: revisionary studies and comparative geography. Bulletin of the American Museum of Natural History 162: 267— SiS:
Sakagami, S. F. 1954. Ueber einige Hummelarten von Hokkaido und Kurilen Inseln (systematische Studien der Hummeln. V). Kontyit 21: 84-92.
— 1972. Bumble bees collected by the California Academy — Lingnan Dawn-Redwood Expedition to central west China, 1948. Pan-Pacific Entomologist 48: 153-174.
CHECKLIST OF BUMBLE BEES
1976. Specific differences in the bionomic characters of bum- blebees. A comparative review. Journal of the Faculty of Science, Hokkaido University (Zoology) 20: 390-447.
Sakagami, S. F., Akahira, Y. & Zucchi, R. 1967. Nest architecture and brood development in a Neotropical bumblebee, Bombus atratus. Insectes sociaux 14: 389-414.
Sakagami, S. F. & Ishikawa, R. 1969. Note préliminaire sur la répartition géographique des bourdons japonais, avec descriptions et remarques sur quelques formes nouvelles ou peu connues. Journal of the Faculty of Science, Hokkaido University (Zoology) 17: 152-196.
1972. Note supplémentaire sur la taxonomie et répartition géographique de quelques bourdons Japonais, avec la description d’une nouvelle sous-espéce. Bulletin of the National Science Mu- seum, Tokyo 15: 607-616.
Sakagami, S. F. & Yoshikawa, K. 1961. Bees of Xylocopinae and Apinae collected by the Osaka City University Expedition to Southeast Asia 1957-58, with some biological notes. Nature and Life in Southeast Asia 1: 409-444.
Sakagami, S. F. & Zucchi, R. 1965. Winterverhalten einer neotropischen Hummel, Bombus atratus, innerhalb des Beobachtungskastens. Ein Beitrag zur Biologie der Hummeln. Journal of the Faculty of Science, Hokkaido University (Zoology) 15: 712-762.
Sandhouse, G. A. 1943. The type species of the genera and subgenera of bees. Proceedings of the United States National Museum 92: 519-619.
Saunders, E. 1896. The Hymenoptera Aculeata of the British Islands. A descriptive account of the families, genera, and species indig- enous to Great Britain and Ireland, with notes as to habits, localities, habitats, etc. viiit391 pp. London.
Say, T. 1837. Descriptions of new species of North American Hy- menoptera, and observations on some already described. Boston Journal of Natural History 1: 361-416.
Schenck, A. 1859. Die nassauischen Bienen. Revision und Erganzung der friiheren Bearbeitungen. Jahrbuch des Nassauischen Vereins fiir Naturkunde 14: 1-414.
Schmiedeknecht, H. L. O. 1878. Monographie der in Thiiringen vorkommenden Arten der Hymenopteren-Gattung Bombus mit einer allgemeinen Einleitung in dieses Genus. Jenaische Zeitschrift fiir Naturwissenschaft 12: 303-430.
1883. Apidae Europaeae (Die Bienen Europa’s) per genera, species et varietates. Dispositae atque descriptae. Volume | fascicule 6, pp. 9(393)-29(413), 1(415)-56(470). Berlin.
Scholl, A. & Obrecht, E. 1983. Enzymelektrophoretische Untersuchungen zurArtabgrenzung imBombus lucorum—Komplex (Apidae, Bombini). Apidologie 14: 65-78.
Scholl, A., Obrecht, E. & Owen, R. E. 1990. The genetic relation- ship between Bombus moderatus Cresson and the Bombus lucorum auct. species complex (Hymenoptera: Apidae). Canadian Journal of Zoology 68: 2264-2268.
Scholl, A., Obrecht, E. & Zimmermann, M. 1992. Evidence of hybridization of Bombus argillaceus and B. ruderatus (Hymenop- tera: Apidae) in a zone of contact in France: enzyme electrophoretic data. Proceedings of the XIX International Congress of Entomol- ogy, Beijing p. 53.
Scholl, A., Thorp, R. W. & Obrecht, E. 1992. The genetic relation- ship between Bombus franklini (Frison) and other taxa of the subgenus Bombus s.str. (Hymenoptera: Apidae). Pan-Pacific Ento- mologist 68: 46-51.
Scholl, A., Thorp, R. W., Bishop, J. A. & Obrecht, E. 1995. The taxonomic status of Bombus alboanalis Franklin and its relation- ship with other taxa of the subgenus Pyrobombus from North America and Europe (Hymenoptera: Apidae). Pan-Pacific Ento- mologist 71: 113-120.
Scholl, A., Thorp, R. W., Owen, R. E., & Obrecht, E. 1992. Specific distinctness of Bombus nevadensis (Cresson) and B. auricomus (Robertson) (Hymenoptera: Apidae) — enzyme electrophoretic data. Journal of the Kansas Entomological Society 65: 134-140.
Schonherr, C. J. 1809. Entomologiska anmarkningar och
147
beskrifningar pa nagra for Svenska fauna nya insecter. Kungliga Svenska vetenskapsakademiens handlingar 30: 48-58.
Schrank, F. de P. 1781. Enumeratio insectorum Austriae indigen- orum. 548 pp. Vienna.
1802. Fauna boica. Durchgedachte Geschichte der in Baiern einheimischen und zahmen Thiere. 412 pp. Ingolstadt.
Schulthess-Rechberg, V. 1886. Zur Hummelfauna Corsica’s. Mitteilungen der Schweizerischen entomologischen Gesellschaft 7: 272-2717.
Schulz, W. A. 1906. Spolia hymenopterologica. 356 pp. Paderborn.
Schwartz, M., Gusenleitner, F., Westrich, P. & Dathe, H. H. 1996. Katalog der Bienen Osterreichs, Deutschlands, und der Schweiz (Hymenoptera, Apidae). Entomofauna (supplement) 8: 398 pp.
Scopoli, J. A. 1763. Entomologia Carniolica exhibens insecta Carniolie indigena et distributa in ordines, genera, species, varietates. 420 pp. Vindobonae.
1777. Introductio ad historiam naturalem sistens genera
lapidum, plantarum, et animalium hactenus detecta, caracteribus essentialibus donata, in tribus divisa, subinde ad leges naturae. 506 pp. Pragae.
Seidl, W. B. 1837. Die in B6hmen vorkommenden Hummelarten. Beitrdge zur gesammten Natur- und Heilwissenschaft 2: 65-73. Sherborn, C. D. 1925. /ndex animalium sive index nominum quae ab A.D. MDCCLVIII generibus et speciebus animalium imposita sunt.
Part VI. Index ceyl.-concolor. pp. 1197-1452. London.
Sianturi, E. M. T., Sota, T., Kato, M., Salmah, S. & Dahelmi 1995. Nest structure, colony composition and foraging activity of a tropical-montane bumblebee, Bombus senex (Hymenoptera: Apidae), in West Sumatra. Japanese Journal of Entomology 63: 657-667.
Sichel, J. 1862. Observations hyménoptérologiques [sic]. Annales de la Société entomologique de France (4) 2: 119-122.
Silveira, F A. & Cure, J. R. 1993. High-altitude bee fauna of southeastern Brazil: implications for biogeographic patterns (Hy- menoptera: Apoidea). Studies on Neotropical Fauna and Environment 28: 47-55.
Skorikov, A. S. 1910a. [Bombus mendax Gerst. and its variations (Hymenoptera, Bombidae).] Russkoe éntomologicheskoe obozrénie 9(1909): 328-330.
1910b. [Nouvelles formes des bourdons (Hymenoptera,
Bombidae). (Diagnoses préliminaires).] Ill. Russkoe
éntomologicheskoe obozrénie 9(1909): 409-413.
1910c. Revision der in der Sammlung des weil. Prof. E. A. Eversmann befindlichen Hummeln. 7rudy Russkago éntomo- logicheskago obshchestva 39: 570-584.
— 1912. Neuve Hummelformen (Hymenoptera, Bombidae). IV. Russkoe éntomologicheskoe obozrénie 12: 606-610.
— 1913. Neue Hummelformen (Hymenoptera, Bombidae). V. Russkoe éntomologicheskoe obozrénie 13: 171-175.
1914a. Les formes nouvelles des bourdons (Hymenoptera,
Bombidae). VI. Russkoe éntomologicheskoe obozrénie 14: 119-129.
1914b. Pratobombus leucopygos [sic] (F. Mor.) et ses variations (Hymenoptera, Bombidae). Russkoe éntomologicheskoe obozrénie 14: 293-294.
— 1914c. Contribution a la faune des bourdons de la partie méridionale de la province Maritime. Russkoe éntomologicheskoe obozrénie 14: 398-407.
1922a. [Palaearctic bumble bees. Part I. General biology (in-
cluding zoogeography).] IJzvestiya Severnoi oblastnoi stantsit
zashchity rastenii ot vreditelei 4: 1-160.
1922b. [Fauna of the Petrograd Province. Bumble bees of the
Petrograd Province.] Faunae Petropolitanae catalogus. Pertro-
gradskii agronomicheskii institut 2(11): 51 pp.
1925. Neue Hummel-Formen (Hymenoptera, Bombidae), VII.
Russkoe éntomologicheskoe obozrénie 19: 115-118.
1931. Die Hummelfauna Turkestans und ihre Beziehungen zur
zentralasiatischen Fauna (Hymenoptera, Bombidae).Abhandlungen
der Pamir-Expedition 1928 8: 175-247.
1933a. Zur Hummelfauna Japans und seiner Nachbarlander.
Mushi 6: 53-65.
148
— 1933b. Zur Fauna und Zoogeographie der Hummeln des Himalaya. Doklady Akademii nauk SSSR 5: 243-248.
—— 1937. Die grénlandischen Hummeln im Aspekte der Zirkumpolarfauna. Entomologiske Meddelelser 20: 37-64.
—— 1938a. Zoogeographische Gesetzmiassigkeiten der Hummelfauna im Kaukasus, Iran und Anatolien (Hymenoptera, Bombinae). Entomologicheskoe obozrenie 27: 145-151.
1938b. Vorlaufige Mitteilung tiber die Hummelfauna Burmas. Arkiv for zoologi 30B: 1-3.
Sladen, F. W. L. 1919. The wasps and bees collected by the Canadian Arctic Expedition, 1913-18. In: Report of the Canadian Arctic Expedition, 1913-18. Volume III: Insects, pp. 25G—35G.
Smith, F. 1844. Notes on the British humble-bees (Bombus of au- thors). Zoologist 2: 541-550.
1849. Notes. Zoologist 7: ix.
1852a. Descriptions of some new and apparently undescribed
species of hymenopterous insects from north China, collected by
Robert Fortune, Esq. Transactions of the Entomological Society of
London 2: 33-45.
1852b. Descriptions of some hymenopterous insects from north-
ern India. Transactions of the Entomological Society of London 2:
45-48.
1854. Catalogue of hymenopterous insects in the collection of
the British Museum. Part II. Apidae. 266(199-465) pp. London.
1861. Descriptions of new genera and species of exotic Hy-
menoptera. Journal of Entomology 1: 146-155.
1869. Descriptions of Hymenoptera from Japan. Entomologist
4: 205-208.
1871. Descriptions of some new insects collected by Dr. Anderson
during the expedition to Yunan [sic]. Proceedings of the Zoological
Society of London 1871: 244-249.
1873. Descriptions of aculeate Hymenoptera of Japan, collected
by Mr. George Lewis at Nagasaki and Hiogo. Transactions of the
Entomological Society of London 1873: 181-206.
1878. List of the Hymenoptera obtained by Mr Ossian Limborg
east of Maulmain, Tenasserim Provinces, during the months of
December 1876, January, March and April 1877, with descriptions
of new species. Journal of the Asiatic Society of Bengal 47: 167—
169,
1879. Descriptions of new species of Hymenoptera in the collection of the British Museum. xxi+240 pp. London.
Snow, D. W. 1997. Should the biological be superseded by the phylogenetic species concept? Bulletin of the British Ornitholo- gists’ Club 117: 110-121.
Sparre-Schneider, J. 1909. Hymenoptera aculeata im arktischen Norwegen. Troms@ museums arshefter 29(1906): 81-160.
1918. Die Hummeln der Kristiana-Gegend. Troms@ museums arshefter 40(1917): 1-45.
Spinola, M. 1806. Insectorum Ligurie species nove aut rariores,
quas in agro Ligustico nuper detexit, descripsit. et iconibus illustravit. Tom. 1. xvii+159 pp. Genue.
Splitter, L. J. 1982. Natural kinds and biological species. 289 pp. D.Phil. thesis, University of Oxford.
Starr, C. K. 1989. Bombus folsomi and the origin of Philippine bumble bees (Hymenoptera: Apidae). Systematic Entomology 14: 411-415.
1992. The bumble bees (Hymenoptera: Apidae) of Taiwan. Bulletin of the National Museum of Natural Science 3: 139-157.
Stephen, W. P. 1957. Bumble bees of western America (Hymenop- tera: Apoidea). Technical Bulletin, Oregon State College, Agricultural Experiment Station 40: 163 pp.
Stephens, J. F. 1829. A systematic catalogue of British insects: being an attempt to arrange all the hitherto discovered indigenous in- sects in accordance with their natural affinities. xxxiv+388 pp. London.
1846. Illustrations of British entomology; or a synopsis of indigenous insects: containing their generic and specific distinc- tions; with an account of their metamorphoses, times of appearance, localities, food, and economy, as far as practicable. Supplement. vi+32 pp. London.
P.H. WILLIAMS
Stevens, O. A. 1948. Native bees. Bumblebees. Bimonthly Bulletin, North Dakota Agricultural Experiment Station 11: 49-54.
Svensson, B. G. 1973. Morphological studies on the two Scandinavian subspecies of Bombus lapponicus Fabricius (Hym. Apidae). Entomologisk tidskrift 94: 140-147.
1979. Pyrobombus lapponicus auct., in Europe recognized as two species: P. lapponicus (Fabricius, 1793) and P. monticola (Smith, 1849) (Hymenoptera, Apoidea, Bombinae). Entomologica scandinavica 10; 275-296.
Svensson, B. G. & Bergstrom, G. 1977. Volatile marking secretions from the labial gland of north European Pyrobombus D. T. males (Hymenoptera, Apidae). Insectes Sociaux 24: 213-224.
Svensson, B. G. & Lundberg, H. 1977. Distribution of bumble bee nests in a subalpine/alpine area in relation to altitude and habitat (Hymenoptera, Apidae). Zoon 5: 63-72.
Swederus, N. S. 1787. Fortsattning af beskrifningen pa 50 nya species af insecter. Kungliga Svenska vetenskapsakademiens handlingar 8: 276-290.
Templeton, A. R. 1981. Mechanisms of speciation — a population genetic approach. Annual Review of Ecology and Systematics 12: 23-48.
Tennessen, K. J. 1997. The rate of species descriptions in Odonata. Entomological news 108: 122-126.
Thomson, C. G. 1872. Hymenoptera scandinaviae. (Apis Lin.). 2: 286 pp. Lund.
Thorp, R. W. 1962. Notes on the distributions of some bumblebees of western North America (Hymenoptera: Apidae). Pan-Pacific Ento- mologist 38: 21-28.
1969. The identity of Bombus vandykei (Hymenoptera: Apidae).
Pan-Pacific Entomologist 45: 87-96.
1970. The type locality of Bombus franklini and notes on putative Arizona records of other Bombini (Hymenoptera: Apidae). Pan-Pacific Entomologist 46: 177-180.
Thorp, R. W., Horning, D. S. & Dunning, L. L. 1983. Bumble bees and cuckoo bumble bees of California (Hymenoptera: Apidae). Bulletin of the California Insect Survey 23: viiit+79 pp.
Tkalci, B. 1959. Eine neue Art der Gattung Psithyrus Lep. aus Korea (Hymenoptera, Bombinae). Casopis Ceskoslovenské spolecnosti entomologické 56: 251-254.
1960. Remerques sur quelques espéces de bourdons de Chine
(Hymenoptera, Bombinae). Bulletin de la Société entomologique
de Mulhouse 1960: 66-71.
196la. Zwei chinesische Hummel-Arten (Hymenoptera,
Bombinae). Casopis Ceskoslovenské spolecnosti entomologické
58: 45-59.
1961b. Zur Hummelfauna der Umgebung Kuku-Nors (Hy-
menoptera, Bombinae). Casopis Ceskoslovenské spolecnosti
entomologické 58: 344-379.
1962. Contribution a l'étude des bourdons du Japon (I) (Hy-
menoptera, Apoidea). Bulletin de la Société entomologique de
Mulhouse 1962: 81-100.
1963. Eine neue Hummel-Art der Gattung Agrobombus Vogt aus
dem Alpengebiet (Hymenoptera, Apoidea).CasopisCeskoslovenské spolecnosti entomologické 60: 183-196.
1965. Contribution a |’étude des bourdons du Japon (II) (Hy- menoptera, Apoidea). Bulletin de la Société entomologique de Mulhouse 1965: 1-14.
1966. Megabombus (Fervidobombus) abditus sp. n. aus aquatorial-A frika. Casopis Moravského musea v Brne 51: 271— 274.
1968a. Neue Arten der Unterfamilie Bombinae der palaarkt- ischen Region (Hymenoptera, Apoidea). Sbornik Entomologického oddeéleni Narodnitho musea v Praze 65: 21-51.
1968b. Revision der vier sympatrischen, homochrome geographische Rassen bildenden Hummelarten SO-Asiens (Hy- menoptera, Apoidea, Bombinae). Annotationes Zoologicae et Botanicae 52: \-31.
1969. Ergebnisse der Albanien-Expedition 1961 des Deutschen Entomologischen Institutes. 78. Beitrag. Hymenoptera: Apidae IV (Bombinae). Beitrtige zur Entomologie 19: 887-916.
CHECKLIST OF BUMBLE BEES
— 1972. Arguments contre l’interprétation traditionelle de la phylogénie des abeilles (Hymenoptera, Apoidea). Premiere partie, introduction et exposés fondamentaux. Bulletin de la Société entomologique de Mulhouse 1972: 17-28.
1973. Taxonomie von Pyrobombus brodmannicus (Vogt) (Hy-
menoptera, Apoidea, Bombinae). Acta entomologica bohemo-
slovaca 70: 259-268.
1974a. Ergebnisse der |. und 2. mongolisch-tschechoslowak- ischen entomologisch-botanischen Expedition in der Mongolei. Nr. 29: Hymenoptera, Apoidea, Bombinae. Sbornik faunistickych practi Entomologického oddéleni Narodntho musea v Praze 15: 25— oT:
— 1974b. Eine Hummel-Ausbeute aus dem Nepal-Himalaya (In- secta, Hymenoptera, Apoidea, Bombinae). Senckenbergiana biologica 55: 311— 349.
1975. Beitrag zur Kenntnis der Hummelfauna der franzsischen
Basses-Alpes (Hymenoptera, Apoidea, Bombinae). Sbornik
Slovenského Narodného Muzea 20(1974): 167-186.
1977. Taxonomisches Notizen zu einigen palaarktischen Biene-
narten (Hymenoptera: Apoidea). Véstnik Ceskoslovenské spoleé-
nosti zoologické 41: 223-239.
1987. Nouveaux synonymes chez les Bombinae (Hymenoptera,
Apoidea). Bulletin de la Société entomologique de Mulhouse 1987:
59-64.
1989. NeueTaxa asiatischer Hummeln (Hymenoptera, Apoidea). Acta entomologica bohemoslovaca 86: 39-60.
Vachal, J. 1904. Voyage de M. G. A. Baer au Tucuman (Argentine). Hymenoptera mellifera (Familia unica: Apide). Revue d’entomo- logie 23: 9-26.
Varela, G. 1992. Nota preliminar sobre la fenologia del nido de Bombus bellicosus Smith, 1879 (Hymenoptera, Apoidea). Boletin de la Sociedad zoologica del Uruguay (2) 7: 53-56.
Viereck, H. L. 1904. Synopsis of bees of Oregon, Washington, British Columbia and Vancouver. Canadian Entomologist 36: 93— 100.
Vogt, O. 1908. Bombi (Hummeln). In: Wissenschaftliche Ergebnisse von Expedition Filchner nach China und Tibet 1903-1905, X. Band —I. Teil, 1. Abschnitt: Zoologische Sammlungen, pp. 100-101. Berlin.
— 1909. Studien iiber das Artproblem. |. Mitteilung. Uber das Variieren der Hummeln. 1. Teil. Sitzungsberichte der Gesellschaft naturforschender Freunde zu Berlin 1909: 28-84.
1911. Studien tiber das Artproblem. 2. Mitteilung. Uber das Variieren der Hummeln. 2. Teil. (Schluss). Sitzungsberichte der Gesellschaft naturforschender Freunde zu Berlin 1911: 31-74.
Vollenhoyen, S. C. Snellen van 1873. Description d'un Bombus nouveau de l’isle de Sumatra. Tijdschrift voor entomologie 16: 229-230.
Wahlberg, P. F. 1854. lakttagelser och anmarkningar 6fver de nordiska humlorna. Ofversigt af K. Vetenskapakademiens forhandlingar 11: 199-211.
Walker, M. D. 1995. Patterns and causes of Arctic plant community diversity. In F. S. I. Chapin & C. K6rner: Arctic and alpine biodiversity, pp. 3-20. Berlin.
Wang, S.-f. 1979. Three new species of bumble bees from Tibet. Acta entomologica sinica 22: 188-191.
1982. Hymenoptera: Apidae — Bombus. In: Insects of Xizang.
Volume II, pp. 427-447.
1985. Apidae — Bombus. In: [Organisms of the Tumuefeng
region of Tienshan], pp. 160-165.
1987. Bombus. In F.-s. Huang: Forest insects of Yunnan, pp.
1378-1381. Yunnan.
1988. Hymenoptera: Apidae — genus Bombus. In F.-s. Huang: Insects of Mt. Namjagbarwa region of Xizang, pp. 553-557. Beijing.
Warncke, K. 1986. Die Wildbienen Mitteleuropas ihre giiltigen Namen und thre Verbreitung (Insecta: Hymenoptera). Entomofauna (supplement) 3: 1-128.
White, A. 1851. Note on the natural history of Shetland. Proceedings
149
of the Linnean Society of London 2: 157-158.
Williams, P. H. 1985a. On the distribution of bumble bees (Hy- menoptera, Apidae) with particular regard to patterns within the British Isles. 180 pp. Ph.D. thesis, University of Cambridge.
1985b. A preliminary cladistic investigation of relationships
among the bumble bees (Hymenoptera, Apidae). Systematic Ento-
mology 10: 239- 255.
1985c. Nuclearbombus fallout [Reply to Menke & Carpenter,
1984]. Sphecos 10: 13.
1988. Habitat use by bumble bees (Bombus spp.). Ecological
Entomology 13: 223-237.
1989. Bumble bees — and their decline in Britain. 15 pp. Uford.
1991. The bumble bees of the Kashmir Himalaya (Hymenop-
tera: Apidae, Bombini). Bulletin of the British Museum (Natural
History) (Entomology) 60: 1-204.
1993. Measuring more of biodiversity for choosing conserva-
tion areas, using taxonomic relatedness. /n T.-Y. Moon (ed.):
International Symposium on Biodiversity and Conservation (KEI),
pp. 194-227. Seoul.
1995. Phylogenetic relationships among bumble bees (Bombus Latr.): a reappraisal of morphological evidence. Systematic Ento- mology 19(1994): 327-344.
—— 1996a. Biodiversity value and taxonomic relatedness. /n M. E. Hochberg, J. Clobert & R. Barbault (eds): The genesis and mainte- nance of biological diversity, pp. 261-277. Oxford.
—— 1996b. Mapping variations in the strength and breadth of biogeographic transition zones using species turnover. Proceed- ings of the Royal Society of London (B) 263: 579-S88.
Williams, P. H. & Cameron, S. A. 1993. Bumble bee (Bombus Latr.) records from the Valley of Flowers, Uttar Pradesh. Bulletin of Entomology 31: 125-127.
Williams, P. H. & Humphries, C. J. 1996. Comparing character diversity among biotas. /n K. J. Gaston (ed.): Biodiversity, a biology of numbers and difference, pp. 54-76. Oxford.
Williams, P. H. & Seddon, A. 1993. Mongolian bumble bee survey. In S. Belbin (ed.): Raleigh International Mongolia 92D expedi- tion: June-August 1992 preliminary scientific reports. 6 pp. London.
Yao, J. & Luo, C. 1997. Hymenoptera: Apidae: Bombini. /n X. Yang: Insects of the Three Gorge reservoir area of Yangtze River, pp. 1686-1696. Chongqing.
Yarrow, I. H. H. 1955. Bombus (Alpinobombus) hyperboreus clydensis, n. ssp., from Baffin Island, North-West Territories, Canada. Annals and Magazine of Natural History (12) 8: 151-152.
1967. On the Formicidae of the Azores. Boletim do Museu
municipal do Funchal 21; 24-32.
1968. Kirby’s species of British bees: designation of holotypes and selection of lectotypes. Part |. Introduction and the species of Apis Linnaeus now included in the genera Bombus Latreille and Psithyrus Lepeletier. Proceedings of the Royal Entomological Society of London (B) 37: 9-15.
— 1970. Is Bombus inexspectatus (Tkalcii) a workerless obligate parasite? (Hym. Apidae). /nsectes Sociaux 17: 95-112.
1971. The author and date of certain subgeneric names in
Bombus Latreille (Hymenoptera: Apidae). Journal of Entomology
(B) 40: 27-29.
1978. Notes on British bumblebees. Entomologist’s Monthly Magazine 113: 15-16.
Yasumatsu, K. 1934. Eine neue, Bombus ignitus ahnliche Schmarotzerhummel aus Korea (Hymenoptera, Bombidz). Annotationes zoologicae japonensis 14: 399-403.
1940. Contributions to the hymenopterous fauna of Inner Mon- golia and North China. Transactions of the Sapporo Natural History Society 16: 90-95.
Zeuner, F. E. & Manning, F. J. 1976. A monograph on fossil bees (Hymenoptera: Apoidea). Bulletin of the British Museum (Natural History) (Geology) 27: 151-268.
Zink, R. M. 1996. Bird species diversity. Nature, London 381: 566.
— 1997. Species concepts. Bulletin of the British Ornithologists’ Club 117: 97-109.
150
The index includes references to names of bumble bees in the list, but not to those in the introduction or in the comments on each
INDEX
P. H. WILLIAMS
species. Valid names are shown in bold. Names in the genus group are shown in capitals.
abditus 114
abnormis 124 acutisquameus 104 adventor 107 ADVENTORIBOMBUS 107 affinis 130
afghanus 99
agnatus 126 AGROBOMBUS 107 agrorum 109, 118 alagesianus 134 alboanalis 125 alboniger 128 albopleuralis 115 ALLOPSITHYRUS 102 ALPIGENOBOMBUS 122 alpigenus 122 alpiniformis 120 ALPINOBOMBUS 119 alpinus 120
altaicus 99 americanorum 111 amurensis 121 anachoreta 108 andamanus 128 andreae 125
angustus 123 ANODONTOBOMBUS 123 APATHUS 102 apollineus 131 appositus 122
arcticus 110, 119, 120 ardens 125
arenicola 109 argillaceus 117 armeniacus 105, 118 arvensis 118
ashtoni 104 ASHTONIPSITHYRUS 102 asiaticus 135 assamensis 106 asturiensis 116
atratus 102, 112 atripes 110 ATROCINCTOBOMBUS 128 atrocinctus 129 atropygus 115
audax 130
auricomus 100 avanus 124 avinoviellus 99 azureus 112
baeri 138 baguionensis 125 baichengensis 135 baikalensis 125 balteatus 119 bannitus 107 barbutellus 104 beaticola 125 bellardii \05 bellicosus 112 bicolor 103 bicoloratus 114 bifarius 128 bimaculatus 128
birmanus 106
biroi 126 bischoffiellus 123 bohemicus 104 BOMBELLUS 99 BOMBIAS 100 BOMBUS 99, 129 BOOPOBOMBUS 100 borealis 122 braccatus 106
BRACHYCEPHALIBOMBUS 137
brachycephalus 138 branickii 105 brasiliensis 106, 113 BREMUS 99 breviceps 123 brevivillus 114 brodmannicus 125 bureschi 109 burjaeticus 130 butteli 136
cajennensis 112
calidus 102, 124 californicus 110 caliginosus 127 callophenax 99 campestris 105 canariensis 130
canus 104
carbonarius 112
cardui 118
cascadensis 127 caucasicus 134
celticus 107
centralis 127 CERATOPSITHYRUS 102 charharensis 121 chayaensis 122 chinensis 100, 104, 133 chinganicus 104, 134 chloronotus 105 CHROMOBOMBUS 107 cingulatus 125 cinnameus 107 CITRINOPSITHYRUS 102 citrinus 103, 106 clydensis 119 COCCINEOBOMBUS 138 coccineus 138
cognatus 110
combai 109 CONFUSIBOMBUS 101 confusus 101 consobrinus 115 consultus 103
contiguus 103 controversus 132 convexus 100
coreanus 104
cornutus 104
crawfordi 103
crotchii 136 CROTCHIIBOMBUS 135 cryptarum 130
CULLUMANOBOMBUS 131
cullumanus 131
ezerskii 117
daghestanicus 109 dahlbomii 111 DASYBOMBUS 139 decoomani 104
defector 99
dentatus 123 derhamellus 109 deuteronymus 108 diabolicus 120 difficillimus 121 DIGRESSOBOMBUS 110 digressus 113
diligens 112 distinguendus 121 DIVERSOBOMBUS 115 diversus 115 dolichocephalus 112 duanjiaoris 132 dumoucheli 110
ecuadorius 136 edwardsii 127 elegans 121 elisabethae 105 emiliae 112 EOPSITHYRUS 102 ephippiatus 128 eriophorus 134 erzurumensis 134 eurythorax 133 eversmanni 102, 125
EVERSMANNIBOMBUS 102
eversmanniellus 102 excellens 111
exil 107
exilis 107 EXILOBOMBUS 107 eximius 129 expolitus 104
exul 107
fedtschenkoi 121 ferganicus 105 fernaldae 105
FERNALDAEPSITHYRUS 102
fernaldi 128
ferrugifer 106 FERVIDOBOMBUS 110 fervidus 110 FESTIVOBOMBUS 128 festivus 101, 129 filchnerae 107 flavescens 110, 125 flavicollis 135
flavidus 106
flavifrons 126 flaviventris 134 flavodorsalis 111 flavopilosus 124 flavothoracicus 105, 133 flavus 105, 125 fletcheri 124
florilegus 130
folsomi 128 formosellus 133
CHECKLIST OF BUMBLE BEES
formosus 128
fragrans 121
franklini 129 FRATERNOBOMBUS 135 fraternus 135
frigidus \26
friseanus 133 fulvescens 108 fulvofasciatus 107 FUNEBRIBOMBUS 137 funebris 137 funerarius \06
gansuensis 106 genalis |23 genitalis 124 gerstaeckeri 115 gilgitensis 129 gilvus 109 globosus 106 grahami 123 griseocollis 137 grossiventris 132 guatemalensis 103
haematurus 124 haemorrhoidalis 106 handlirschi 139 handlirschianus 100 haueri 138
hedini 104, 109 heicens 135 himalayanus 99 hispanicus 116 hoenei 104, 133 honshuensis 109 HORTOBOMBUS 115 hortorum 1|16 hortulanus 137 huangcens 135 humilis 108 hummeli 115
huntii 128 hydrophthalmus 132 hyperboreus 119 HYPNOROBOMBUS 124 hypnorum 124 hypocrita \30
ignitus 130 imitator 110 impatiens 128 impetuosus 109 incertus 134 indicus 105 inexspectatus 109 infirmus 124 infrequens 124 insularis 103, 124 interruptus 103, 105 intrudens 103 irisanensis 114
Jacobsoni 130 japonicus 130 Jonellus 125
KALLOBOMBUS 118 kashmirensis \22 keriensis 134 kirbiellus 119 kirbyellus 119 klapperichi 104, 125 kohistanensis 124
kohli 112, 134 konakovi 109 Koreanus 117 koropokkrus 124 kotzschi 126 kozlovi 134 KOZLOVIBOMBUS 132 krusemani 138 kuani 106 Kulingensis 114 kurilensis 116
LABORIOPSITHYRUS 102 laboriosus 113 ladakhensis \33 LAESOBOMBUS 106 laesus 106
laevis 107 LAPIDARIOBOMBUS 132 lapidarius 134 LAPPONICOBOMBUS 124 lapponicus |27 latefasciatus 104 lateralis 128 latissimus 129 latofasciatus 99 lefebvrei 118 lemniscatus \24 lepidus 124 LEUCOBOMBUS 129 leucopygus 124 leucurus 124
licenti 104 liepetterseni 107 ligusticus 117
lit 107
linguarius 115 lissonurus 106 longiceps 135 longipes 115 lucorum 130 lugubris 100, 104, 128 luteipes |24 lutescens 109
macgregori \39 maculidorsis 106 maderensis 130 magnus 130
makarjini 99
malaisei 115, 129 margreiteri 99
martensi: 104 marussinus 99 MASTRUCATOBOMBUS 122 mastrucatus 122 maxillosus 104 maxwelli 115
mearnsi 125
medius 113 MEGABOMBUS 115 melaleucus \36 MELANOBOMBUS 132 melanopoda \\7 melanopygus \27 melanurus 121 menchuae 139 MENDACIBOMBUS 99 mendax 99 meridionalis 106, 116 mesomelas 118 METAPSITHYRUS 102 metcalfi 106 mexicanus 114
mimeticus 115 miniatocaudatus 118, 135 miniatus 133
mirus 124
mixtus 104, 127 mlokosievitzii 109 mocsaryi 106 moderatus 130 modestus 114, 125, 130 mongol 121 monozonus 104, 130 montanus 109, 128 monticola 128 montivagus 115 montivolanoides 125 montivolans 107 morawitzi 104, 132 morawitzianus 105 morawitziides 122 morio \12
morrisoni 137 MUCIDOBOMBUS 101 mucidus \02 muscorum \07 mysticus 103
napensis 138 nasutus 123 nemorum 104, 121 neoboreus 120 neotropicus 138 nepalensis 105 NEVADENSIBOMBUS 100 nevadensis 100 niger 112
nigripes 107, 111 nigriventris 112 nigrodorsalis 113 nigrothoracicus 136 nikiforuki 125 nivalis 119 niveatus 135 NOBILIBOMBUS 122 nobilis \22 norvegicus 105 notocastaneus 117 novus \05
nursei 126 nymphae 125
oberti \32 OBERTOBOMBUS 32 obtusus \34 occidentalis 131 oceanicus |25 ochraceus 105 ochrobasis 134 oculatus 133 ODONTOBOMBUS 115 opifex 112
opulentus 108 orichalceus 123 ORIENTALIBOMBUS 106 orientalis 106
ornatus 128
pallidus 107, 111 paradoxus 101 parthenius \24 pascuorum 109 patagiatus 130 pekingensis 117 pensylvanicus \11
151
152
peralpinus 124 perezi 104, 109 pereziellus 107 perniger 117 perplexus 118, 124 persicus \02 personatus 121 phariensis 133 picipes 125
pieli 105
pleuralis 126 POECILOBOMBUS 123 polaris 120 POMOBOMBUS 118 pomorum 118 portchinsky 116 potanini 109 praticola 127 pratorum \25 PRESSIBOMBUS 129 pressus 129 pretiosus 123 priscus 106
proteus 118 prshewalskyi 133 przewalskiellus 116 pseudobaicalensis 109 PSITHYRUS 102 pulcherrimus 122 pullatus 112 pyramideus 104 pyrenaeus 105, 126 pyrenes 99 PYROBOMBUS 123 pyrosoma 133
quadricolor 106
radoszkowskyi 118 redikorzevi 105
regeli 135
reinigi 130
reinigiellus 116
religiosus 118
remotus 109
reticulatus 133 RHODOBOMBUS 118 richardsi 104, 129, 134 richardsiellus 133 roborowskyi 121 ROBUSTOBOMBUS 136 robustus 137
rohweri 137
rotundiceps 125 RUBICUNDOBOMBUS 138 rubicundus 138 rubriventris 112 RUDERARIOBOMBUS 107 ruderarius 109
ruderatus 117 RUFIPEDIBOMBUS 129 rufipes 129 RUFOCINCTOBOMBUS 131 rufocinctus 131, 133 rufocognitus 123 rufofasciatus 133 rufoflavus 125
rupestris 105
saltuarius 116 saltuum 121 sandersoni 126 scandinavicus 128 schrencki 109 securus 118 semenovi 132 semenovianus 1|33 semenoviellus |32 semialbopleuralis 107 senex 114 SENEXIBOMBUS 114 senilis 108 separandus 134 SEPARATOBOMBUS 137 separatus 137 serrisquama 131 shaposhnikovi 100 shillongensis 125 SIBIRICOBOMBUS 134 sibiricus 134 sichelii 134
signifer 124
sikkimi 122 silantjewi 131 simillimus 132 simulatilis 109 simulus 123 sitkensis 127 skorikovi 106 smithianus 110 sololensis 103 sonani 124
sonomae 110 sonorus 111 SOROEENSIBOMBUS 118 soroeensis 118 sporadicus \29 steindachneri 1\13 stenothorax 133 stramineus 122 strenuus 120, 134 subbaicalensis 108 subdistinctus 121 SUBTERRANEOBOMBUS subterraneus 121 subtypicus 124 suckleyi 104 SULCOBOMBUS 101 sulfureus 135 superbus 99 superequester 109 supremus 115 sushkini 116 susterai 104, 105 susteraianus 105 sylvarum 109 sylvestris 106 sylvicola 127
tahanensis 125 tajushanensis 105, 114 TANGUTICOBOMBUS 132 tanguticus 132
tenellus 134
tenuifasciatus 126, 134 terminalis 129, 130 ternarius 128 TERRESTRIBOMBUS 129
P. H. WILLIAMS
terrestris 130 terricola 131
tersatus 115 tetrachromus 122, 124 thoracicus 109, 112 THORACOBOMBUS 107 tianschanicus 106 tibetanus 104, 124 tibetensis 134 tichenkoi 115
tonsus 133 transbaicalicus 105 transversalis \12 tricolor 119 TRICORNIBOMBUS 110 tricornis 110 trifasciatus 115 trilineatus 134 trinominatus 114 tristis 108, 119 tucumanus 137 tunicatus 129 turkestanicus 99 turneri 104
UNCOBOMBUS 124 unicolor 104, 109 unicus 132 ussurensis 115
vagans 127 validus 122 vallestris 129 vandykei 127 variabilis 103, 108 variopictus 133 varius 99, 105 vasilievi 130 velox 109 velutinus 112 vestalis 104 veteranus 109 villarricaensis 117 violaceus 112 vogti 136 vogtiellus 109 VOLUCELLOBOMBUS 136 volucelloides 136 vorticosus 135 vosnesenskit 128
waltoni 100 waterstoni 133 weisit 113 wilemani 115 wilmattae 128 wurflenii 122 wutaishanensis 133
xelajuensis 114 xionglaris 132 xizangensis 122
yezoensis 116
yuennanensis 109 yuennanicola 124 yuennanicus 118
zhadaensis 132 zhaosu 127 zonatus 108
Bulletin of The Natural History Museum Entomology Series
Earlier Entomology Bulletins are still in print. The following can be ordered from Intercept (address on inside front cover). Where the complete backlist is not shown, this may also be obtained from the same address.
Volume 58
No. | The mealybug genus Planococcus (Homoptera: Pseudococcidae). J.M. Cox. 1989. Pp. 1-78, 40 figs.
No. 2 The Simuliidae (Diptera) of the Santiago onchocerciasis focus of Ecuador. A.J. Shelley, M. Arzube & C.A. Couch. 1989. Pp. 79-130, 153 figs (including 2 plates in colour).
Volume 59
No. 1 The songs of the western European bush-crickets of the genus Platycleis in relation to their taxonomy (Orthoptera: Tettigoniidae). D.R. Ragge. 1990. Pp. 1-35. A reclassification of the Melanotus group of genera (Coleoptera: Elateridae). C.M.F. von Hayek. 1990. Pp. 37-115.
No. 2 The green lacewings of the world: a generic review (Neuroptera: Chrysopidae). S.J. Brooks & P.C. Barnard. 1990. Pp. 117-286.
Volume 60 No. | The bumble bees of the Kashmir Himalaya (Hymenoptera: Apidae, Bombini). P.H. Williams. 1991. Pp. 1-204.
No. 2 Sattleria: a European genus of brachypterous alpine moths (Lepidoptera: Gelechiidae). L.M. Pitkin & K. Sattler. 1991. Pp. 205-241. A review of wing reduction in Lepidoptera. K. Sattler. 1991. Pp. 243-288.
Volume 61 No. | Thrips (Thysanoptera) from Pakistan to the Pacific: a review. J.M. Palmer. 1992. Pp. 1-76. No. 2 Neotropical red-brown Ennominae in the genera Thysanopyga Herrich-Scaffer and Perissopteryx Warren
(Lepidoptera: Geometridae). M. Kruger & M.J. Scoble. 1992. Pp. 77-148.
Volume 62
No. | Caloptilia \eaf-miner moths (Gracillariidae) of South-East Asia. Decheng Yuan and Gaden S. Robinson. 1993. Pp. 1-37.
No. 2 Neotropical Emerald moths of the genera Nemoria, Lissochlora and Chavarriella, with particular reference to the species of Costa Rica (Lepidoptera: Geometridae, Geometrinae). Linda M. Pitkin. 1993. Pp. 39-159.
Volume 63
No. | A revision of the Indo-Pacific species of Ooencyrtus (Hymenoptera: Encyrtidae), parasitoids of the
immature stages of economically important insect species (mainly Hemiptera and Lepidoptera). D.W. Huang and J.S. Noyes. 1994. Pp. 1-135.
No. 2 A taxonomic review of the common green lacewing genus Chrysoperla (Neuroptera: Chrysopidae). S.J. Brooks. 1994. Pp. 137-210.
Volume 64
No. 1 Revision of the neotropical genus Oospila Warren (Lepidoptera: Geometridae) M.A. Cook and M.J. Scoble. 1995. Pp. 1-115.
No. 2 Encyrtidae of Costa Rica (Hymenhoptera: Chalcidoidea): the genus Aenasius Walker, parasitoids of mealybugs (Homoptera: Pseudococcidae). J.S. Noyes and H. Ren. 1995. Pp. 117-164.
Volume 65
No. | A revised classification of the Asian and Pacific selenocephaline leafhoppers (Homoptera: Cicadellidae). Y. Zhang and M.D. Webb. 1996. Pp 1-103.
No. 2 Encyrtidae (Hymenoptera: Chalcidoidea) of Costa Rica: the genera and species associated with jumping
plant-lice (Homoptera: Psylloidea). J.S. Noyes and P. Hanson. 1996. Pp. 105-164.
Volume 66
No. | A revised classification of the Asian and Pacific selenocephaline leafhoppers (Homoptera: Cicadellidae). Y. Zhang and M.D. Webb. 1997. Pp 1-121.
No. 2 Microtermes in East Africa (Isoptera: Termitidae: Macrotermitinae) S. Bacchus. 1997. Pp. 123-171.
Mealybugs of the genera Eumyrmococcus Silvestri and Xenococcus — Silvestri associated with the ant genus Acropyga Roger and a review
_ of the subfamily Rhizoecinae (Hemiptera, Coccoidea, Pseudococeidas D.J. Williams Monophyly of the dacetonine tribe-group and its component tribes (Hymenoptera: Formicidae) Barry Bolton An annotated checklist of bumble bees with an analysis of patterns se description (Hymenoptera: Apidae, Bombini)
Paul H. Williams
ENTOMOLOGY SERIES